Skip to content

Use --find-links for MLIR-AIE and LLVM-AIE wheel sources#107

Open
x4v13r64 wants to merge 3 commits into
amd:develfrom
x4v13r64:fix/pin-package-indexes
Open

Use --find-links for MLIR-AIE and LLVM-AIE wheel sources#107
x4v13r64 wants to merge 3 commits into
amd:develfrom
x4v13r64:fix/pin-package-indexes

Conversation

@x4v13r64
Copy link
Copy Markdown

@x4v13r64 x4v13r64 commented Apr 14, 2026

The GitHub Release URLs for mlir_aie and llvm-aie are currently added as --extra-index-url entries alongside PyPI. This means pip queries all indexes for all packages, which can cause resolution issues when the same package name exists on multiple indexes at different versions. Switching to --find-links adds the wheel files as candidates without creating a competing package index.

@x4v13r64
Use --find-links for MLIR-AIE and LLVM-AIE wheel sources
be7084b 
Replace --extra-index-url with --find-links for the GitHub Release
wheel sources. This avoids adding them as full package indexes that
pip queries for all packages. The CPU-only PyTorch index is kept as
an --extra-index-url since it serves a standard PEP 503 index.
@github-actions
Copy link
Copy Markdown
Contributor

github-actions Bot commented May 20, 2026 •
edited
Loading

CI Test Results

05eb076 (2026_05_20_03_31_19)

IRON - CI Summary

Examples

iron/applications/llama_3.2_1b
Test Krackan Status Krackan Phoenix Status Phoenix
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_1] ✅ - - -
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_40] ✅ - - -
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_1] ✅ - - -
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_40] ✅ - - -

Small

iron/operators/axpy
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0] ✅ 164.40 ✅ 330.12
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0] ✅ 202.22 ✅ 549.88
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0] ✅ 172.40 ✅ 334.56
test_axpy[input_length_2048-num_aie_columns_8-tile_size_256-scalar_factor_3.0] ✅ 171.54 - -
iron/operators/dequant
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32] ✅ 181.72 ✅ 349.90
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32] ✅ 201.86 ✅ 461.30
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32] ✅ 172.90 ✅ 392.72
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32] ✅ 184.24 ✅ 473.38
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32] ✅ 171.88 ✅ 547.34
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32] ✅ 180.24 ✅ 455.44
test_dequant[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32] ✅ 182.72 - -
test_dequant[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32] ✅ 258.46 - -
iron/operators/elementwise_add
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048] ✅ 137.68 ✅ 341.80
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024] ✅ 154.76 ✅ 437.28
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512] ✅ 167.88 ✅ 488.16
test_elementwise_add[input_length_2048-num_aie_columns_8-tile_size_256] ✅ 222.26 - -
iron/operators/elementwise_mul
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048] ✅ 152.32 ✅ 263.60
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024] ✅ 188.30 ✅ 483.68
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512] ✅ 174.46 ✅ 388.34
test_elementwise_mul[input_length_2048-num_aie_columns_8-tile_size_256] ✅ 229.64 - -
iron/operators/gelu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] ✅ 176.68 ✅ 387.72
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] ✅ 163.70 ✅ 321.16
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] ✅ 196.08 ✅ 307.54
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] ✅ 183.60 ✅ 456.32
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] ✅ 185.54 ✅ 442.78
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] ✅ 240.12 ✅ 442.44
test_gelu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] ✅ 215.28 - -
test_gelu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] ✅ 247.98 - -
iron/operators/gemm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_gemm[M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1] ✅ 2439.50 - -
test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1] ✅ 229.42 ✅ 722.84
test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1] ✅ 236.50 ✅ 545.60
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1] ✅ 48815.42 ✅ 84194.24
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1] ✅ 28794.04 ✅ 25421.12
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1] ✅ 7909.12 - -
test_gemm[M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1] ✅ 2293.32 ✅ 4790.38
test_gemm[M_64-K_512-N_256-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_16-k_64-n_64-trace_size_0-partition_N_4] ✅ 3695.02 ✅ 5724.56
test_gemm[M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1] ✅ 1235.48 - -
iron/operators/gemv
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128] ✅ 0.21 ✅ 0.11
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048] ✅ 12.08 ✅ 3.69
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024] ✅ 24.18 ✅ 6.39
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512] ✅ 39.42 ✅ 8.40
test_gemv[M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256] ✅ 42.07 - -
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024] ✅ 12.72 ✅ 3.59
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024] ✅ 21.30 ✅ 5.78
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024] ✅ 39.54 ✅ 9.70
test_gemv[M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024] ✅ 40.50 - -
iron/operators/layer_norm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] ✅ 161.50 ✅ 306.20
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] ✅ 162.16 ✅ 333.28
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] ✅ 179.70 ✅ 469.14
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] ✅ 175.30 ✅ 431.90
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] ✅ 160.62 ✅ 446.72
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] ✅ 160.60 ✅ 423.60
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] ✅ 171.42 - -
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] ✅ 245.38 - -
iron/operators/mem_copy
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048] ✅ 196.78 ✅ 434.96
test_mem_copy[input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128] ✅ 228.70 - -
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024] ✅ 182.86 ✅ 320.28
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024] ✅ 224.12 ✅ 391.18
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512] ✅ 184.02 ✅ 514.28
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512] ✅ 180.12 ✅ 487.04
test_mem_copy[input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256] ✅ 198.90 - -
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256] ✅ 221.72 ✅ 513.60
iron/operators/mha
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_mha[seq_len_16384-dim_64-num_heads_1-num_pipelines_8-num_kv_heads_0] ✅ 40505.30 - -
iron/operators/relu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] ✅ 157.34 ✅ 337.02
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] ✅ 222.00 ✅ 481.78
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] ✅ 168.66 ✅ 545.92
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] ✅ 174.10 ✅ 389.70
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] ✅ 186.46 ✅ 379.38
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] ✅ 197.28 ✅ 464.52
test_relu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] ✅ 182.02 - -
test_relu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] ✅ 233.16 - -
iron/operators/rms_norm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False] ✅ 202.88 ✅ 356.96
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True] ✅ 174.96 ✅ 266.38
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False] ✅ 191.90 ✅ 493.98
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True] ✅ 186.26 ✅ 460.18
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False] ✅ 155.02 ✅ 409.42
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True] ✅ 184.68 ✅ 429.52
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False] ✅ 178.30 ✅ 410.04
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True] ✅ 212.08 ✅ 405.64
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False] ✅ 178.44 ✅ 352.06
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True] ✅ 170.68 ✅ 357.38
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False] ✅ 179.04 ✅ 480.80
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_True] ✅ 220.78 - -
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False] ✅ 211.84 - -
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_True] ✅ 225.24 - -
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False] ✅ 216.80 - -
iron/operators/rope
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0] ✅ 158.02 ✅ 294.18
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0] ✅ 215.26 ✅ 408.72
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0] ✅ 199.24 ✅ 374.50
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0] ✅ 155.00 - -
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0] ✅ 172.24 ✅ 369.48
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0] ✅ 174.30 ✅ 443.16
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0] ✅ 160.10 ✅ 433.72
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0] ✅ 172.80 - -
iron/operators/sigmoid
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] ✅ 176.14 ✅ 336.48
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] ✅ 163.14 ✅ 495.30
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] ✅ 160.70 ✅ 357.70
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] ✅ 174.06 ✅ 352.56
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] ✅ 196.88 ✅ 751.56
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] ✅ 191.38 ✅ 537.14
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] ✅ 198.86 - -
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] ✅ 216.70 - -
iron/operators/silu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_silu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] ✅ 182.42 ✅ 324.92
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] ✅ 167.32 ✅ 361.46
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] ✅ 181.02 ✅ 371.90
test_silu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] ✅ 189.86 - -
iron/operators/softmax
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024] ✅ 179.04 ✅ 627.86
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048] ✅ 144.80 ✅ 508.20
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512] ✅ 146.10 ✅ 527.34
iron/operators/swiglu_decode
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584] ✅ 4206.63 ✅ 14924.96
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048] ✅ 4012.70 ✅ 19260.95
iron/operators/swiglu_prefill
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False] ✅ 9107.87 ✅ 24045.09
iron/operators/tanh
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_tanh[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] ✅ 147.40 ✅ 264.02
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] ✅ 156.98 ✅ 798.28
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] ✅ 172.84 ✅ 337.02
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] ✅ 177.58 ✅ 343.92
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] ✅ 166.78 ✅ 314.08
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] ✅ 175.26 ✅ 416.36
test_tanh[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] ✅ 186.06 - -
test_tanh[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] ✅ 194.40 - -
iron/operators/transpose
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8] ✅ 197.48 ✅ 513.56
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8] ✅ 194.06 ✅ 454.26
Krackan - Small

IRON

Tested on 2026_05_20_03_31_19 at commit 05eb076.

iron/operators/axpy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0]✅ 5/5164.400.08n/a
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]✅ 5/5202.220.06n/a
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]✅ 5/5172.400.07n/a
test_axpy[input_length_2048-num_aie_columns_8-tile_size_256-scalar_factor_3.0]✅ 5/5171.540.07n/a
iron/operators/dequant
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32]✅ 5/5181.720.03n/a
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]✅ 5/5201.860.03n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]✅ 5/5172.900.03n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]✅ 5/5184.240.03n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]✅ 5/5171.880.03n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]✅ 5/5180.240.03n/a
test_dequant[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32]✅ 5/5182.720.03n/a
test_dequant[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32]✅ 5/5258.460.02n/a
iron/operators/elementwise_add
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5137.680.09n/a
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5154.760.08n/a
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5167.880.08n/a
test_elementwise_add[input_length_2048-num_aie_columns_8-tile_size_256]✅ 5/5222.260.06n/a
iron/operators/elementwise_mul
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5152.320.09n/a
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5188.300.07n/a
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5174.460.07n/a
test_elementwise_mul[input_length_2048-num_aie_columns_8-tile_size_256]✅ 5/5229.640.06n/a
iron/operators/gelu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5176.680.05n/a
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5163.700.05n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5196.080.04n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5183.600.05n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5185.540.04n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5240.120.04n/a
test_gelu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5215.280.04n/a
test_gelu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5247.980.03n/a
iron/operators/gemm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gemm[M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1]✅ 5/52439.503.931547.23
test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1]✅ 5/5229.421.0042.84
test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1]✅ 5/5236.500.9942.34
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]✅ 5/548815.420.52351.94
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]✅ 5/528794.040.87596.66
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1]✅ 5/57909.123.182173.06
test_gemm[M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1]✅ 5/52293.323.52924.34
test_gemm[M_64-K_512-N_256-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_16-k_64-n_64-trace_size_0-partition_N_4]✅ 5/53695.020.3518.74
test_gemm[M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1]✅ 5/51235.485.591726.50
iron/operators/gemv
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128]✅ 5/5n/a0.210.20
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a12.0812.08
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]✅ 5/5n/a24.1824.17
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]✅ 5/5n/a39.4239.39
test_gemv[M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256]✅ 5/5n/a42.0742.05
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a12.7212.71
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a21.3021.29
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a39.5439.52
test_gemv[M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a40.5040.47
iron/operators/layer_norm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5161.500.05n/a
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5162.160.05n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5179.700.05n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5175.300.05n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5160.620.05n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5160.600.05n/a
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5171.420.05n/a
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5245.380.04n/a
iron/operators/mem_copy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048]✅ 5/5196.780.04n/a
test_mem_copy[input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128]✅ 5/5228.700.04n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]✅ 5/5182.860.05n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]✅ 5/5224.120.04n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]✅ 5/5184.020.05n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]✅ 5/5180.120.05n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256]✅ 5/5198.900.04n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]✅ 5/5221.720.04n/a
iron/operators/mha
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_mha[seq_len_16384-dim_64-num_heads_1-num_pipelines_8-num_kv_heads_0]✅ 5/540505.300.21n/a
iron/operators/relu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5157.340.05n/a
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5222.000.04n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5168.660.05n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5174.100.05n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5186.460.04n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5197.280.04n/a
test_relu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5182.020.05n/a
test_relu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5233.160.04n/a
iron/operators/rms_norm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False]✅ 5/5202.880.04n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]✅ 5/5174.960.07n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]✅ 5/5191.900.04n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]✅ 5/5186.260.06n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]✅ 5/5155.020.05n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]✅ 5/5184.680.06n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]✅ 5/5178.300.05n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]✅ 5/5212.080.05n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]✅ 5/5178.440.05n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]✅ 5/5170.680.06n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]✅ 5/5179.040.05n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_True]✅ 5/5220.780.04n/a
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False]✅ 5/5211.840.04n/a
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_True]✅ 5/5225.240.04n/a
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False]✅ 5/5216.800.04n/a
iron/operators/rope
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0]✅ 5/5158.020.66n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]✅ 5/5215.260.49n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]✅ 5/5199.240.51n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0]✅ 5/5155.000.65n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]✅ 5/5172.240.46n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]✅ 5/5174.300.45n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]✅ 5/5160.100.47n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0]✅ 5/5172.800.44n/a
iron/operators/sigmoid
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5176.140.05n/a
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5163.140.05n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5160.700.05n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5174.060.05n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5196.880.04n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5191.380.04n/a
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5198.860.05n/a
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5216.700.04n/a
iron/operators/silu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_silu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5182.420.05n/a
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5167.320.05n/a
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5181.020.05n/a
test_silu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5189.860.04n/a
iron/operators/softmax
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024]✅ 5/5179.040.75n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]✅ 5/5144.800.93n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5146.100.91n/a
iron/operators/swiglu_decode
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584]✅ 5/54206.630.00n/a
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]✅ 5/54012.700.00n/a
iron/operators/swiglu_prefill
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False]✅ 5/59107.870.24n/a
iron/operators/tanh
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_tanh[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5147.400.06n/a
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5156.980.05n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5172.840.05n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5177.580.05n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5166.780.05n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5175.260.05n/a
test_tanh[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5186.060.04n/a
test_tanh[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5194.400.04n/a
iron/operators/transpose
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8]✅ 5/5197.482.66n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8]✅ 5/5194.062.75n/a

Trends:

IRON Trends

iron/operators/axpy

test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.10 (+23.52%)0.08 (+6.06%)0.07 (-2.35%)0.07 (+8.73%)0.01 (+63.07%)183.10 (-8.04%)164.40 (-4.70%)171.30 (+2.39%)120.90 (-19.08%)24.94 (+17.42%)
5503a95 — 2026-05-12 00:06:190.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)199.10 (n/a)172.50 (n/a)167.30 (n/a)149.40 (n/a)21.24 (n/a)

test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (-26.59%)0.06 (-11.09%)0.07 (+2.22%)0.05 (-7.23%)0.01 (-45.97%)257.80 (+7.78%)202.22 (+9.43%)186.30 (-2.20%)175.40 (+36.18%)34.88 (-20.61%)
5503a95 — 2026-05-12 00:06:190.10 (n/a)0.07 (n/a)0.06 (n/a)0.05 (n/a)0.02 (n/a)239.20 (n/a)184.80 (n/a)190.50 (n/a)128.80 (n/a)43.93 (n/a)

test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.09 (+29.96%)0.07 (+19.41%)0.07 (+13.43%)0.06 (+29.80%)0.01 (+18.43%)202.70 (-22.96%)172.40 (-16.54%)171.90 (-11.85%)141.90 (-23.09%)21.57 (-32.45%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)263.10 (n/a)206.56 (n/a)195.00 (n/a)184.50 (n/a)31.94 (n/a)

test_axpy[input_length_2048-num_aie_columns_8-tile_size_256-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.09 (+9.75%)0.07 (+7.08%)0.07 (+9.16%)0.06 (+12.72%)0.01 (+20.29%)196.40 (-11.25%)171.54 (-6.37%)166.90 (-8.40%)140.90 (-8.92%)24.01 (-1.35%)
5503a95 — 2026-05-12 00:06:190.08 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)221.30 (n/a)183.22 (n/a)182.20 (n/a)154.70 (n/a)24.34 (n/a)
iron/operators/dequant

test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.03 (-18.68%)0.03 (-11.17%)0.03 (-13.39%)0.02 (-10.04%)0.00 (-41.09%)215.80 (+11.12%)181.72 (+11.17%)174.70 (+15.47%)160.40 (+23.01%)21.97 (-21.89%)
5503a95 — 2026-05-12 00:06:190.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)194.20 (n/a)163.46 (n/a)151.30 (n/a)130.40 (n/a)28.13 (n/a)

test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (-1.53%)0.03 (-7.87%)0.03 (-0.42%)0.01 (-45.88%)0.01 (+81.77%)355.90 (+84.79%)201.86 (+18.50%)177.60 (+0.40%)143.00 (+1.56%)87.40 (+259.02%)
5503a95 — 2026-05-12 00:06:190.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)192.60 (n/a)170.34 (n/a)176.90 (n/a)140.80 (n/a)24.34 (n/a)

test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.03 (-26.06%)0.03 (-8.47%)0.03 (+2.36%)0.02 (-8.35%)0.00 (-50.32%)212.50 (+9.14%)172.90 (+6.74%)158.70 (-2.28%)157.20 (+35.28%)23.69 (-28.47%)
5503a95 — 2026-05-12 00:06:190.05 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)194.70 (n/a)161.98 (n/a)162.40 (n/a)116.20 (n/a)33.12 (n/a)

test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (+25.93%)0.03 (+5.15%)0.03 (+1.00%)0.02 (-6.16%)0.01 (+107.30%)224.00 (+6.57%)184.24 (-1.96%)191.60 (-0.98%)121.60 (-20.57%)38.02 (+67.27%)
5503a95 — 2026-05-12 00:06:190.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)210.20 (n/a)187.92 (n/a)193.50 (n/a)153.10 (n/a)22.73 (n/a)

test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (+6.48%)0.03 (+5.92%)0.03 (+3.25%)0.03 (+8.86%)0.00 (-7.03%)208.10 (-8.16%)171.88 (-6.22%)172.70 (-3.14%)137.90 (-6.13%)26.14 (-20.21%)
5503a95 — 2026-05-12 00:06:190.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)226.60 (n/a)183.28 (n/a)178.30 (n/a)146.90 (n/a)32.76 (n/a)

test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (+5.87%)0.03 (+6.12%)0.03 (+7.21%)0.03 (+3.57%)0.00 (+6.00%)200.60 (-3.46%)180.24 (-5.75%)188.10 (-6.70%)138.00 (-5.54%)24.50 (-4.65%)
5503a95 — 2026-05-12 00:06:190.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)207.80 (n/a)191.24 (n/a)201.60 (n/a)146.10 (n/a)25.69 (n/a)

test_dequant[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.03 (-8.09%)0.03 (-5.54%)0.03 (-15.01%)0.03 (+4.68%)0.00 (-38.94%)196.40 (-4.47%)182.72 (+4.44%)190.00 (+17.65%)154.40 (+8.81%)17.65 (-39.03%)
5503a95 — 2026-05-12 00:06:190.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)205.60 (n/a)174.96 (n/a)161.50 (n/a)141.90 (n/a)28.95 (n/a)

test_dequant[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.03 (+0.44%)0.02 (-14.52%)0.02 (-9.66%)0.02 (-29.09%)0.01 (+144.53%)329.10 (+41.00%)258.46 (+22.62%)236.30 (+10.68%)182.60 (-0.44%)66.22 (+266.93%)
5503a95 — 2026-05-12 00:06:190.03 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)233.40 (n/a)210.78 (n/a)213.50 (n/a)183.40 (n/a)18.05 (n/a)
iron/operators/elementwise_add

test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.11 (n/a)0.09 (n/a)0.09 (n/a)0.08 (n/a)0.01 (n/a)157.10 (n/a)137.68 (n/a)139.30 (n/a)115.90 (n/a)15.49 (n/a)

test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.10 (n/a)0.08 (n/a)0.08 (n/a)0.07 (n/a)0.01 (n/a)177.40 (n/a)154.76 (n/a)162.60 (n/a)124.10 (n/a)22.35 (n/a)

test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.09 (n/a)0.08 (n/a)0.08 (n/a)0.05 (n/a)0.01 (n/a)228.20 (n/a)167.88 (n/a)158.80 (n/a)139.70 (n/a)36.03 (n/a)

test_elementwise_add[input_length_2048-num_aie_columns_8-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.08 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.02 (n/a)311.40 (n/a)222.26 (n/a)215.80 (n/a)148.40 (n/a)63.32 (n/a)
iron/operators/elementwise_mul

test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.13 (n/a)0.09 (n/a)0.07 (n/a)0.07 (n/a)0.02 (n/a)188.50 (n/a)152.32 (n/a)164.00 (n/a)98.00 (n/a)35.11 (n/a)

test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.10 (n/a)0.07 (n/a)0.07 (n/a)0.05 (n/a)0.02 (n/a)260.20 (n/a)188.30 (n/a)183.20 (n/a)129.20 (n/a)47.16 (n/a)

test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.10 (n/a)0.07 (n/a)0.07 (n/a)0.06 (n/a)0.02 (n/a)208.40 (n/a)174.46 (n/a)182.50 (n/a)126.10 (n/a)36.44 (n/a)

test_elementwise_mul[input_length_2048-num_aie_columns_8-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.09 (n/a)0.06 (n/a)0.05 (n/a)0.04 (n/a)0.03 (n/a)337.40 (n/a)229.64 (n/a)226.90 (n/a)138.20 (n/a)91.88 (n/a)
iron/operators/gelu

test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (n/a)0.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)220.60 (n/a)176.68 (n/a)184.40 (n/a)109.90 (n/a)47.23 (n/a)

test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.00 (n/a)180.30 (n/a)163.70 (n/a)164.00 (n/a)144.20 (n/a)13.64 (n/a)

test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)219.50 (n/a)196.08 (n/a)196.40 (n/a)171.40 (n/a)20.81 (n/a)

test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)213.90 (n/a)183.60 (n/a)182.00 (n/a)164.90 (n/a)19.50 (n/a)

test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.00 (n/a)204.00 (n/a)185.54 (n/a)189.60 (n/a)156.30 (n/a)19.19 (n/a)

test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)314.00 (n/a)240.12 (n/a)210.10 (n/a)166.30 (n/a)66.34 (n/a)

test_gelu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)275.80 (n/a)215.28 (n/a)204.60 (n/a)178.80 (n/a)36.42 (n/a)

test_gelu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)312.60 (n/a)247.98 (n/a)241.70 (n/a)219.80 (n/a)37.64 (n/a)
iron/operators/gemm

test_gemm[M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:555.03 (-2.04%)3.93 (-15.30%)3.73 (-18.64%)3.43 (-17.77%)0.66 (+62.84%)2743.00 (+21.61%)2439.50 (+19.72%)2518.60 (+22.91%)1869.50 (+2.07%)361.99 (+103.37%)1978.80 (-2.04%)1547.23 (-15.30%)1468.85 (-18.64%)1348.67 (-17.77%)260.69 (+62.84%)
5503a95 — 2026-05-12 00:06:195.13 (n/a)4.64 (n/a)4.59 (n/a)4.17 (n/a)0.41 (n/a)2255.50 (n/a)2037.64 (n/a)2049.10 (n/a)1831.50 (n/a)177.99 (n/a)2019.91 (n/a)1826.68 (n/a)1805.34 (n/a)1640.13 (n/a)160.08 (n/a)

test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:551.34 (+8.42%)1.00 (-10.43%)0.98 (-15.64%)0.71 (-18.68%)0.23 (+55.75%)309.70 (+22.99%)229.42 (+14.49%)226.80 (+18.50%)164.50 (-7.74%)52.22 (+74.60%)57.38 (+8.42%)42.84 (-10.43%)41.60 (-15.64%)30.47 (-18.68%)9.65 (+55.75%)
5503a95 — 2026-05-12 00:06:191.24 (n/a)1.12 (n/a)1.16 (n/a)0.88 (n/a)0.15 (n/a)251.80 (n/a)200.38 (n/a)191.40 (n/a)178.30 (n/a)29.91 (n/a)52.92 (n/a)47.83 (n/a)49.32 (n/a)37.47 (n/a)6.19 (n/a)

test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:551.31 (-8.48%)0.99 (-17.62%)1.02 (-18.30%)0.62 (-24.42%)0.25 (+0.27%)354.80 (+32.34%)236.50 (+23.74%)217.30 (+22.42%)168.60 (+9.27%)70.22 (+51.45%)55.97 (-8.48%)42.34 (-17.62%)43.42 (-18.30%)26.60 (-24.42%)10.55 (+0.27%)
5503a95 — 2026-05-12 00:06:191.43 (n/a)1.20 (n/a)1.25 (n/a)0.82 (n/a)0.25 (n/a)268.10 (n/a)191.12 (n/a)177.50 (n/a)154.30 (n/a)46.37 (n/a)61.16 (n/a)51.39 (n/a)53.15 (n/a)35.20 (n/a)10.52 (n/a)

test_gemm[M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:550.52 (+0.04%)0.52 (-0.02%)0.52 (-0.02%)0.51 (-0.09%)0.00 (+75.60%)48900.90 (+0.09%)48815.42 (+0.02%)48817.20 (+0.02%)48742.60 (-0.04%)57.02 (+75.65%)352.46 (+0.04%)351.94 (-0.02%)351.92 (-0.02%)351.32 (-0.09%)0.41 (+75.61%)
5503a95 — 2026-05-12 00:06:190.52 (n/a)0.52 (n/a)0.52 (n/a)0.52 (n/a)0.00 (n/a)48854.90 (n/a)48807.76 (n/a)48807.50 (n/a)48763.20 (n/a)32.46 (n/a)352.31 (n/a)351.99 (n/a)351.99 (n/a)351.65 (n/a)0.23 (n/a)

test_gemm[M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:550.88 (-1.65%)0.87 (-0.64%)0.87 (-1.03%)0.87 (+0.41%)0.00 (-63.86%)28899.10 (-0.41%)28794.04 (+0.63%)28846.60 (+1.04%)28568.60 (+1.68%)131.27 (-63.44%)601.36 (-1.65%)596.66 (-0.64%)595.56 (-1.03%)594.48 (+0.41%)2.73 (-63.86%)
5503a95 — 2026-05-12 00:06:190.90 (n/a)0.88 (n/a)0.88 (n/a)0.87 (n/a)0.01 (n/a)29017.60 (n/a)28613.74 (n/a)28550.40 (n/a)28096.10 (n/a)359.06 (n/a)611.47 (n/a)600.48 (n/a)601.74 (n/a)592.05 (n/a)7.56 (n/a)

test_gemm[M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:553.29 (-2.29%)3.18 (-1.81%)3.17 (-0.29%)3.09 (-2.54%)0.07 (-19.63%)8143.40 (+2.61%)7909.12 (+1.83%)7928.00 (+0.29%)7660.40 (+2.35%)179.93 (-15.88%)2242.70 (-2.29%)2173.06 (-1.81%)2166.99 (-0.29%)2109.67 (-2.54%)49.60 (-19.63%)
5503a95 — 2026-05-12 00:06:193.36 (n/a)3.24 (n/a)3.18 (n/a)3.17 (n/a)0.09 (n/a)7936.30 (n/a)7767.18 (n/a)7905.10 (n/a)7484.70 (n/a)213.91 (n/a)2295.35 (n/a)2213.21 (n/a)2173.25 (n/a)2164.71 (n/a)61.71 (n/a)

test_gemm[M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:553.80 (-11.03%)3.52 (-13.65%)3.53 (-15.93%)3.23 (-9.66%)0.21 (-28.57%)2496.20 (+10.70%)2293.32 (+15.61%)2284.30 (+18.95%)2120.90 (+12.40%)136.08 (-11.57%)996.70 (-11.03%)924.34 (-13.65%)925.41 (-15.93%)846.86 (-9.66%)54.13 (-28.57%)
5503a95 — 2026-05-12 00:06:194.27 (n/a)4.08 (n/a)4.20 (n/a)3.57 (n/a)0.29 (n/a)2255.00 (n/a)1983.64 (n/a)1920.40 (n/a)1887.00 (n/a)153.89 (n/a)1120.28 (n/a)1070.40 (n/a)1100.77 (n/a)937.46 (n/a)75.78 (n/a)

test_gemm[M_64-K_512-N_256-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_16-k_64-n_64-trace_size_0-partition_N_4]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:550.48 (-6.85%)0.35 (-3.29%)0.32 (-4.35%)0.29 (-3.36%)0.08 (-14.33%)4282.10 (+3.47%)3695.02 (+2.65%)3898.20 (+4.55%)2587.90 (+7.36%)654.63 (-5.57%)25.93 (-6.85%)18.74 (-3.29%)17.22 (-4.35%)15.67 (-3.36%)4.12 (-14.33%)
5503a95 — 2026-05-12 00:06:190.52 (n/a)0.36 (n/a)0.33 (n/a)0.30 (n/a)0.09 (n/a)4138.30 (n/a)3599.66 (n/a)3728.60 (n/a)2410.60 (n/a)693.24 (n/a)27.84 (n/a)19.38 (n/a)18.00 (n/a)16.22 (n/a)4.81 (n/a)

test_gemm[M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:556.63 (+0.84%)5.59 (+10.28%)6.13 (+24.75%)4.01 (+6.62%)1.13 (+10.06%)1658.20 (-6.21%)1235.48 (-8.93%)1086.00 (-19.84%)1004.00 (-0.83%)280.44 (+1.52%)2047.03 (+0.84%)1726.50 (+10.28%)1892.52 (+24.75%)1239.45 (+6.62%)348.92 (+10.06%)
5503a95 — 2026-05-12 00:06:196.57 (n/a)5.07 (n/a)4.91 (n/a)3.76 (n/a)1.03 (n/a)1768.00 (n/a)1356.64 (n/a)1354.80 (n/a)1012.40 (n/a)276.26 (n/a)2030.00 (n/a)1565.55 (n/a)1517.00 (n/a)1162.48 (n/a)317.03 (n/a)
iron/operators/gemv

test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:550.26 (+14.68%)0.21 (+6.09%)0.20 (-1.47%)0.19 (+32.99%)0.03 (-12.93%)0.25 (+14.68%)0.20 (+6.09%)0.19 (-1.47%)0.18 (+32.99%)0.03 (-12.93%)
5503a95 — 2026-05-12 00:06:190.23 (n/a)0.19 (n/a)0.20 (n/a)0.14 (n/a)0.03 (n/a)0.22 (n/a)0.19 (n/a)0.20 (n/a)0.14 (n/a)0.03 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5513.23 (+0.76%)12.08 (-5.06%)12.32 (-3.42%)10.69 (-12.34%)1.18 (+242.61%)13.22 (+0.76%)12.08 (-5.06%)12.31 (-3.42%)10.68 (-12.34%)1.18 (+242.61%)
5503a95 — 2026-05-12 00:06:1913.13 (n/a)12.73 (n/a)12.75 (n/a)12.19 (n/a)0.35 (n/a)13.12 (n/a)12.72 (n/a)12.74 (n/a)12.18 (n/a)0.34 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5524.69 (-2.82%)24.18 (-0.92%)24.01 (-1.90%)23.94 (+2.05%)0.31 (-55.19%)24.67 (-2.82%)24.17 (-0.92%)23.99 (-1.90%)23.92 (+2.05%)0.31 (-55.19%)
5503a95 — 2026-05-12 00:06:1925.41 (n/a)24.41 (n/a)24.47 (n/a)23.46 (n/a)0.70 (n/a)25.39 (n/a)24.39 (n/a)24.46 (n/a)23.44 (n/a)0.70 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5541.47 (+2.18%)39.42 (-0.59%)39.00 (-1.43%)37.93 (-1.68%)1.36 (+75.69%)41.44 (+2.18%)39.39 (-0.59%)38.97 (-1.43%)37.91 (-1.68%)1.36 (+75.69%)
5503a95 — 2026-05-12 00:06:1940.58 (n/a)39.65 (n/a)39.56 (n/a)38.58 (n/a)0.77 (n/a)40.56 (n/a)39.63 (n/a)39.54 (n/a)38.56 (n/a)0.77 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5543.79 (-1.31%)42.07 (-3.25%)41.94 (-3.51%)39.91 (-6.59%)1.50 (+155.58%)43.76 (-1.31%)42.05 (-3.25%)41.92 (-3.51%)39.88 (-6.59%)1.50 (+155.58%)
5503a95 — 2026-05-12 00:06:1944.37 (n/a)43.48 (n/a)43.47 (n/a)42.72 (n/a)0.59 (n/a)44.34 (n/a)43.46 (n/a)43.44 (n/a)42.70 (n/a)0.59 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5513.22 (-1.53%)12.72 (+0.56%)12.77 (-2.86%)12.03 (+11.41%)0.51 (-51.91%)13.22 (-1.53%)12.71 (+0.56%)12.77 (-2.86%)12.02 (+11.41%)0.51 (-51.91%)
5503a95 — 2026-05-12 00:06:1913.43 (n/a)12.65 (n/a)13.15 (n/a)10.80 (n/a)1.07 (n/a)13.42 (n/a)12.64 (n/a)13.14 (n/a)10.79 (n/a)1.07 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5524.22 (-0.92%)21.30 (-10.95%)23.80 (-1.27%)16.34 (-28.60%)3.85 (+504.20%)24.20 (-0.92%)21.29 (-10.95%)23.78 (-1.27%)16.33 (-28.60%)3.84 (+504.20%)
5503a95 — 2026-05-12 00:06:1924.44 (n/a)23.92 (n/a)24.10 (n/a)22.89 (n/a)0.64 (n/a)24.43 (n/a)23.90 (n/a)24.09 (n/a)22.87 (n/a)0.64 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5541.14 (+0.13%)39.54 (-0.17%)38.67 (-3.50%)38.26 (+1.94%)1.42 (-9.48%)41.12 (+0.13%)39.52 (-0.17%)38.65 (-3.50%)38.24 (+1.94%)1.42 (-9.48%)
5503a95 — 2026-05-12 00:06:1941.09 (n/a)39.61 (n/a)40.08 (n/a)37.53 (n/a)1.57 (n/a)41.06 (n/a)39.59 (n/a)40.05 (n/a)37.51 (n/a)1.57 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5542.94 (-6.87%)40.50 (-3.39%)41.90 (+2.10%)35.63 (-8.49%)2.93 (+6.08%)42.91 (-6.87%)40.47 (-3.39%)41.87 (+2.10%)35.60 (-8.49%)2.93 (+6.08%)
5503a95 — 2026-05-12 00:06:1946.11 (n/a)41.92 (n/a)41.04 (n/a)38.93 (n/a)2.76 (n/a)46.08 (n/a)41.89 (n/a)41.01 (n/a)38.91 (n/a)2.76 (n/a)
iron/operators/layer_norm

test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)211.10 (n/a)161.50 (n/a)146.80 (n/a)134.10 (n/a)31.21 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)236.70 (n/a)162.16 (n/a)158.00 (n/a)114.60 (n/a)45.38 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)210.30 (n/a)179.70 (n/a)179.80 (n/a)135.80 (n/a)28.39 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)227.60 (n/a)175.30 (n/a)160.10 (n/a)142.10 (n/a)38.04 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.00 (n/a)169.40 (n/a)160.62 (n/a)166.70 (n/a)137.90 (n/a)13.27 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)184.40 (n/a)160.60 (n/a)169.60 (n/a)127.60 (n/a)21.91 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)222.40 (n/a)171.42 (n/a)163.70 (n/a)134.10 (n/a)37.94 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)312.00 (n/a)245.38 (n/a)223.20 (n/a)177.50 (n/a)60.73 (n/a)
iron/operators/mem_copy

test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (-4.91%)0.04 (+2.76%)0.04 (-5.17%)0.04 (+49.28%)0.01 (-43.74%)232.20 (-33.01%)196.78 (-8.34%)207.40 (+5.49%)167.70 (+5.14%)28.24 (-62.95%)
5503a95 — 2026-05-12 00:06:190.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)346.60 (n/a)214.68 (n/a)196.60 (n/a)159.50 (n/a)76.20 (n/a)

test_mem_copy[input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (-17.59%)0.04 (-3.93%)0.04 (-4.48%)0.03 (+45.11%)0.00 (-83.15%)240.80 (-31.08%)228.70 (-1.41%)231.90 (+4.70%)219.20 (+21.37%)9.33 (-86.36%)
5503a95 — 2026-05-12 00:06:190.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)349.40 (n/a)231.98 (n/a)221.50 (n/a)180.60 (n/a)68.43 (n/a)

test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-4.21%)0.05 (-10.55%)0.05 (-5.98%)0.03 (-25.55%)0.01 (+81.33%)237.30 (+34.30%)182.86 (+15.95%)174.20 (+6.41%)137.00 (+4.42%)44.18 (+159.19%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.05 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)176.70 (n/a)157.70 (n/a)163.70 (n/a)131.20 (n/a)17.04 (n/a)

test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (-11.78%)0.04 (-20.19%)0.04 (-8.75%)0.03 (-39.15%)0.01 (+82.18%)316.00 (+64.33%)224.12 (+35.08%)183.10 (+9.58%)153.80 (+13.34%)77.14 (+253.06%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)192.30 (n/a)165.92 (n/a)167.10 (n/a)135.70 (n/a)21.85 (n/a)

test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-14.93%)0.05 (-11.27%)0.04 (-14.84%)0.04 (-11.15%)0.01 (-15.48%)215.10 (+12.56%)184.02 (+12.65%)195.40 (+17.43%)143.60 (+17.51%)29.32 (+15.49%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)191.10 (n/a)163.36 (n/a)166.40 (n/a)122.20 (n/a)25.39 (n/a)

test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-3.40%)0.05 (+3.74%)0.05 (+11.95%)0.04 (-8.52%)0.01 (-7.93%)229.40 (+9.29%)180.12 (-3.65%)173.90 (-10.68%)139.80 (+3.56%)32.19 (+8.38%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.04 (n/a)0.04 (n/a)0.04 (n/a)0.01 (n/a)209.90 (n/a)186.94 (n/a)194.70 (n/a)135.00 (n/a)29.70 (n/a)

test_mem_copy[input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (-33.98%)0.04 (-6.34%)0.04 (+14.15%)0.04 (+32.00%)0.00 (-83.73%)211.50 (-24.25%)198.90 (-1.97%)197.20 (-12.39%)182.50 (+51.45%)11.99 (-80.80%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.02 (n/a)279.20 (n/a)202.90 (n/a)225.10 (n/a)120.50 (n/a)62.46 (n/a)

test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.04 (-26.29%)0.04 (-16.01%)0.04 (-12.05%)0.03 (-15.80%)0.01 (-37.69%)281.80 (+18.80%)221.72 (+17.95%)214.50 (+13.67%)195.60 (+35.64%)35.00 (+2.36%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)237.20 (n/a)187.98 (n/a)188.70 (n/a)144.20 (n/a)34.19 (n/a)
iron/operators/mha

test_mha[seq_len_16384-dim_64-num_heads_1-num_pipelines_8-num_kv_heads_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.21 (+0.34%)0.21 (-0.01%)0.21 (+0.12%)0.21 (-0.67%)0.00 (+82.65%)40877.50 (+0.67%)40505.30 (+0.01%)40526.00 (-0.12%)40111.70 (-0.34%)271.92 (+83.27%)
5503a95 — 2026-05-12 00:06:190.21 (n/a)0.21 (n/a)0.21 (n/a)0.21 (n/a)0.00 (n/a)40604.80 (n/a)40502.20 (n/a)40576.30 (n/a)40247.10 (n/a)148.37 (n/a)
iron/operators/rms_norm

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (-22.91%)0.04 (-26.00%)0.04 (-36.74%)0.04 (-20.88%)0.01 (-27.88%)233.00 (+26.36%)202.88 (+34.52%)217.80 (+58.06%)155.80 (+29.73%)31.84 (+14.96%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.06 (n/a)0.06 (n/a)0.04 (n/a)0.01 (n/a)184.40 (n/a)150.82 (n/a)137.80 (n/a)120.10 (n/a)27.70 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.11 (+22.28%)0.07 (+0.45%)0.07 (-5.98%)0.05 (-2.38%)0.02 (+61.38%)240.80 (+2.42%)174.96 (+2.82%)171.10 (+6.41%)114.00 (-18.22%)48.38 (+28.81%)
5503a95 — 2026-05-12 00:06:190.09 (n/a)0.07 (n/a)0.08 (n/a)0.05 (n/a)0.01 (n/a)235.10 (n/a)170.16 (n/a)160.80 (n/a)139.40 (n/a)37.56 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (-34.05%)0.04 (-32.77%)0.05 (-27.46%)0.04 (-28.03%)0.01 (-48.82%)230.50 (+38.94%)191.90 (+47.16%)179.70 (+37.81%)165.10 (+51.61%)26.41 (+11.71%)
5503a95 — 2026-05-12 00:06:190.08 (n/a)0.06 (n/a)0.06 (n/a)0.05 (n/a)0.01 (n/a)165.90 (n/a)130.40 (n/a)130.40 (n/a)108.90 (n/a)23.64 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (-1.38%)0.06 (-11.67%)0.06 (-10.51%)0.04 (-30.93%)0.01 (+111.29%)255.10 (+44.78%)186.26 (+16.66%)175.80 (+11.76%)144.10 (+1.41%)41.75 (+220.43%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.06 (n/a)0.07 (n/a)0.06 (n/a)0.01 (n/a)176.20 (n/a)159.66 (n/a)157.30 (n/a)142.10 (n/a)13.03 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (-7.08%)0.05 (-5.39%)0.05 (-0.36%)0.05 (-2.23%)0.01 (-21.87%)176.20 (+2.26%)155.02 (+4.86%)152.50 (+0.39%)123.70 (+7.66%)21.34 (-14.75%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)172.30 (n/a)147.84 (n/a)151.90 (n/a)114.90 (n/a)25.03 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (-2.02%)0.06 (+2.04%)0.06 (+26.61%)0.04 (-12.45%)0.01 (+7.40%)252.30 (+14.21%)184.68 (-0.85%)162.60 (-21.03%)139.30 (+2.05%)46.48 (+26.11%)
5503a95 — 2026-05-12 00:06:190.08 (n/a)0.06 (n/a)0.05 (n/a)0.05 (n/a)0.01 (n/a)220.90 (n/a)186.26 (n/a)205.90 (n/a)136.50 (n/a)36.86 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-18.04%)0.05 (-7.49%)0.04 (-5.76%)0.04 (+26.88%)0.01 (-52.06%)211.30 (-21.22%)178.30 (-0.44%)190.80 (+6.12%)134.60 (+22.03%)30.50 (-53.03%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.02 (n/a)268.20 (n/a)179.08 (n/a)179.80 (n/a)110.30 (n/a)64.93 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-11.92%)0.05 (+0.65%)0.05 (+12.46%)0.03 (-5.59%)0.01 (-24.00%)329.50 (+5.91%)212.08 (-2.69%)193.70 (-11.07%)153.30 (+13.56%)68.06 (-1.87%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.05 (n/a)0.04 (n/a)0.03 (n/a)0.02 (n/a)311.10 (n/a)217.94 (n/a)217.80 (n/a)135.00 (n/a)69.35 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-3.75%)0.05 (-5.03%)0.04 (-7.80%)0.04 (+16.22%)0.01 (-38.53%)195.70 (-13.94%)178.44 (+2.43%)183.10 (+8.47%)140.20 (+3.93%)22.33 (-44.69%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)227.40 (n/a)174.20 (n/a)168.80 (n/a)134.90 (n/a)40.37 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.07 (-2.20%)0.06 (+14.84%)0.06 (+17.46%)0.05 (+62.05%)0.01 (-45.97%)200.60 (-38.30%)170.68 (-19.75%)159.70 (-14.87%)137.20 (+2.31%)27.07 (-65.33%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.05 (n/a)0.05 (n/a)0.03 (n/a)0.02 (n/a)325.10 (n/a)212.68 (n/a)187.60 (n/a)134.10 (n/a)78.07 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (-26.26%)0.05 (-9.99%)0.05 (-5.21%)0.04 (-9.63%)0.01 (-41.42%)233.00 (+10.64%)179.04 (+9.01%)167.40 (+5.48%)155.90 (+35.57%)31.92 (-10.43%)
5503a95 — 2026-05-12 00:06:190.07 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)210.60 (n/a)164.24 (n/a)158.70 (n/a)115.00 (n/a)35.64 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (-13.20%)0.04 (-0.95%)0.05 (+16.79%)0.03 (-8.12%)0.01 (-6.52%)347.60 (+8.86%)220.78 (+1.61%)183.50 (-14.37%)169.20 (+15.26%)75.38 (+15.45%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.04 (n/a)0.04 (n/a)0.03 (n/a)0.01 (n/a)319.30 (n/a)217.28 (n/a)214.30 (n/a)146.80 (n/a)65.29 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (+5.02%)0.04 (-4.82%)0.04 (-10.13%)0.03 (+3.96%)0.01 (+15.49%)284.50 (-3.79%)211.84 (+5.85%)196.90 (+11.31%)155.80 (-4.77%)55.01 (+1.59%)
5503a95 — 2026-05-12 00:06:190.05 (n/a)0.04 (n/a)0.05 (n/a)0.03 (n/a)0.01 (n/a)295.70 (n/a)200.14 (n/a)176.90 (n/a)163.60 (n/a)54.15 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.06 (-0.77%)0.04 (-20.22%)0.04 (-21.70%)0.03 (-24.62%)0.01 (+55.00%)273.10 (+32.70%)225.24 (+29.81%)230.80 (+27.73%)141.90 (+0.78%)49.84 (+98.67%)
5503a95 — 2026-05-12 00:06:190.06 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.01 (n/a)205.80 (n/a)173.52 (n/a)180.70 (n/a)140.80 (n/a)25.09 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.05 (+17.15%)0.04 (+22.52%)0.04 (+22.25%)0.03 (+13.75%)0.01 (+3.53%)324.10 (-12.10%)216.80 (-19.41%)195.10 (-18.20%)169.30 (-14.62%)63.76 (-20.91%)
5503a95 — 2026-05-12 00:06:190.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)368.70 (n/a)269.00 (n/a)238.50 (n/a)198.30 (n/a)80.61 (n/a)
iron/operators/rope

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.79 (-3.25%)0.66 (+2.26%)0.71 (+0.69%)0.41 (+33.56%)0.15 (-26.13%)239.80 (-25.13%)158.02 (-8.91%)138.20 (-0.65%)123.90 (+3.34%)47.01 (-43.60%)
5503a95 — 2026-05-12 00:06:190.82 (n/a)0.64 (n/a)0.71 (n/a)0.31 (n/a)0.20 (n/a)320.30 (n/a)173.48 (n/a)139.10 (n/a)119.90 (n/a)83.35 (n/a)

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.61 (-30.81%)0.49 (-27.67%)0.52 (-23.56%)0.33 (-32.06%)0.13 (-11.12%)302.40 (+47.22%)215.26 (+41.67%)189.00 (+30.80%)161.40 (+44.49%)62.75 (+82.94%)
5503a95 — 2026-05-12 00:06:190.88 (n/a)0.67 (n/a)0.68 (n/a)0.48 (n/a)0.14 (n/a)205.40 (n/a)151.94 (n/a)144.50 (n/a)111.70 (n/a)34.30 (n/a)

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.63 (-10.18%)0.51 (-19.68%)0.45 (-30.68%)0.41 (-13.82%)0.10 (+10.47%)240.90 (+16.04%)199.24 (+25.74%)216.70 (+44.27%)157.10 (+11.34%)37.65 (+35.55%)
5503a95 — 2026-05-12 00:06:190.70 (n/a)0.63 (n/a)0.65 (n/a)0.47 (n/a)0.09 (n/a)207.60 (n/a)158.46 (n/a)150.20 (n/a)141.10 (n/a)27.78 (n/a)

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.73 (-3.69%)0.65 (+8.40%)0.68 (+25.96%)0.49 (-5.84%)0.10 (-3.29%)200.30 (+6.20%)155.00 (-7.67%)144.40 (-20.57%)134.00 (+3.80%)26.84 (+8.77%)
5503a95 — 2026-05-12 00:06:190.76 (n/a)0.60 (n/a)0.54 (n/a)0.52 (n/a)0.10 (n/a)188.60 (n/a)167.88 (n/a)181.80 (n/a)129.10 (n/a)24.68 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.60 (+0.39%)0.46 (-18.29%)0.45 (-21.86%)0.30 (-37.79%)0.12 (+145.72%)247.60 (+60.78%)172.24 (+29.29%)164.50 (+28.02%)122.40 (-0.41%)49.56 (+291.49%)
5503a95 — 2026-05-12 00:06:190.60 (n/a)0.56 (n/a)0.57 (n/a)0.48 (n/a)0.05 (n/a)154.00 (n/a)133.22 (n/a)128.50 (n/a)122.90 (n/a)12.66 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.57 (+8.09%)0.45 (-1.61%)0.39 (-19.26%)0.32 (+17.33%)0.11 (+7.11%)231.50 (-14.76%)174.30 (+0.65%)187.70 (+23.89%)128.60 (-7.48%)43.78 (-21.69%)
5503a95 — 2026-05-12 00:06:190.53 (n/a)0.45 (n/a)0.49 (n/a)0.27 (n/a)0.11 (n/a)271.60 (n/a)173.18 (n/a)151.50 (n/a)139.00 (n/a)55.91 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.58 (-5.68%)0.47 (-2.82%)0.48 (+3.28%)0.34 (-16.83%)0.09 (+13.60%)214.10 (+20.21%)160.10 (+4.22%)153.40 (-3.16%)126.40 (+5.95%)33.22 (+52.47%)
5503a95 — 2026-05-12 00:06:190.62 (n/a)0.49 (n/a)0.47 (n/a)0.41 (n/a)0.08 (n/a)178.10 (n/a)153.62 (n/a)158.40 (n/a)119.30 (n/a)21.79 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.53 (+3.66%)0.44 (+0.43%)0.46 (-4.95%)0.35 (+29.25%)0.08 (-14.49%)212.10 (-22.62%)172.80 (-2.94%)161.00 (+5.16%)139.20 (-3.53%)34.27 (-37.07%)
5503a95 — 2026-05-12 00:06:190.51 (n/a)0.44 (n/a)0.48 (n/a)0.27 (n/a)0.10 (n/a)274.10 (n/a)178.04 (n/a)153.10 (n/a)144.30 (n/a)54.46 (n/a)
iron/operators/softmax

test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.85 (-20.62%)0.75 (-22.72%)0.77 (-24.14%)0.59 (-30.34%)0.11 (+14.81%)221.10 (+43.57%)179.04 (+30.86%)171.30 (+31.87%)153.40 (+25.94%)29.04 (+101.48%)
5503a95 — 2026-05-12 00:06:191.08 (n/a)0.97 (n/a)1.01 (n/a)0.85 (n/a)0.10 (n/a)154.00 (n/a)136.82 (n/a)129.90 (n/a)121.80 (n/a)14.41 (n/a)

test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:551.15 (+1.87%)0.93 (-1.72%)0.98 (+2.85%)0.72 (-7.47%)0.18 (+26.71%)182.70 (+8.11%)144.80 (+3.05%)133.80 (-2.83%)114.40 (-1.80%)28.51 (+36.07%)
5503a95 — 2026-05-12 00:06:191.12 (n/a)0.95 (n/a)0.95 (n/a)0.78 (n/a)0.14 (n/a)169.00 (n/a)140.52 (n/a)137.70 (n/a)116.50 (n/a)20.96 (n/a)

test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:551.05 (-7.98%)0.91 (-9.58%)0.83 (-18.59%)0.82 (-3.82%)0.11 (+7.97%)159.10 (+3.99%)146.10 (+10.93%)158.20 (+22.92%)124.60 (+8.63%)17.27 (+22.73%)
5503a95 — 2026-05-12 00:06:191.14 (n/a)1.00 (n/a)1.02 (n/a)0.86 (n/a)0.10 (n/a)153.00 (n/a)131.70 (n/a)128.70 (n/a)114.70 (n/a)14.07 (n/a)
iron/operators/swiglu_decode

test_swiglu_decode[embedding_dim_1024-hidden_dim_3584]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.00 (+0.00%)0.00 (-7.41%)0.00 (+0.00%)0.00 (-22.22%)0.00 (+53.39%)5465.52 (+20.30%)4206.63 (+7.06%)3864.11 (+0.32%)3503.42 (+0.53%)834.33 (+78.73%)
5503a95 — 2026-05-12 00:06:190.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)4543.17 (n/a)3929.28 (n/a)3851.61 (n/a)3484.84 (n/a)466.81 (n/a)

test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.00 (+0.00%)0.00 (-2.83%)0.00 (-9.09%)0.00 (+5.56%)0.00 (-5.56%)4310.17 (-4.16%)4012.70 (+3.23%)4163.46 (+10.27%)3561.69 (-0.67%)324.25 (-10.66%)
5503a95 — 2026-05-12 00:06:190.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)4497.33 (n/a)3887.13 (n/a)3775.82 (n/a)3585.56 (n/a)362.95 (n/a)
iron/operators/swiglu_prefill

test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:550.28 (+0.94%)0.24 (+8.61%)0.27 (+4.42%)0.15 (-4.02%)0.06 (-2.71%)14417.41 (+4.22%)9107.87 (-7.77%)7849.62 (-4.21%)7539.34 (-0.93%)2971.49 (+5.16%)
5503a95 — 2026-05-12 00:06:190.28 (n/a)0.23 (n/a)0.26 (n/a)0.15 (n/a)0.06 (n/a)13833.24 (n/a)9874.95 (n/a)8194.32 (n/a)7610.28 (n/a)2825.75 (n/a)
iron/operators/transpose

test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:552.87 (-17.79%)2.66 (-5.32%)2.63 (-0.56%)2.51 (+11.35%)0.16 (-68.04%)208.80 (-10.19%)197.48 (+3.39%)199.30 (+0.55%)182.40 (+21.60%)11.53 (-64.52%)
5503a95 — 2026-05-12 00:06:193.50 (n/a)2.81 (n/a)2.65 (n/a)2.26 (n/a)0.49 (n/a)232.50 (n/a)191.00 (n/a)198.20 (n/a)150.00 (n/a)32.50 (n/a)

test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:553.53 (-10.53%)2.75 (-23.46%)2.55 (-34.32%)2.50 (-12.46%)0.44 (-5.03%)209.90 (+14.20%)194.06 (+30.93%)206.00 (+52.25%)148.40 (+11.75%)26.13 (+20.60%)
5503a95 — 2026-05-12 00:06:193.95 (n/a)3.59 (n/a)3.88 (n/a)2.85 (n/a)0.47 (n/a)183.80 (n/a)148.22 (n/a)135.30 (n/a)132.80 (n/a)21.66 (n/a)
Krackan - Examples

IRON

Tested on 2026_05_20_03_38_45 at commit 05eb076.

iron/applications/llama_3.2_1b
TestChecksTTFT (mean)TPS (mean)
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_1]✅ 5/52.12n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_40]✅ 5/52.164.15
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_1]✅ 5/52.09n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_40]✅ 5/52.084.14

Trends:

IRON Trends

iron/applications/llama_3.2_1b

test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_1]

Commit/Date TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
05eb076 — 2026-05-20 03:32:572.13 (-0.93%)2.12 (-0.33%)2.12 (-0.66%)2.11 (+0.28%)0.01 (-60.13%)
5503a95 — 2026-05-11 23:56:252.15 (n/a)2.13 (n/a)2.13 (n/a)2.11 (n/a)0.02 (n/a)

test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_40]

Commit/Date TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
05eb076 — 2026-05-20 03:32:574.17 (-0.45%)4.15 (-0.47%)4.15 (-0.41%)4.14 (-0.48%)0.01 (+3.49%)2.27 (+1.20%)2.16 (+0.13%)2.13 (-0.23%)2.12 (+0.19%)0.06 (+26.69%)
5503a95 — 2026-05-11 23:56:254.19 (n/a)4.17 (n/a)4.17 (n/a)4.16 (n/a)0.01 (n/a)2.24 (n/a)2.16 (n/a)2.14 (n/a)2.12 (n/a)0.05 (n/a)

test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_1]

Commit/Date TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
05eb076 — 2026-05-20 03:32:572.11 (+0.72%)2.09 (-0.01%)2.09 (-0.14%)2.07 (-0.72%)0.01 (+327.03%)
5503a95 — 2026-05-11 23:56:252.10 (n/a)2.09 (n/a)2.09 (n/a)2.09 (n/a)0.00 (n/a)

test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_40]

Commit/Date TPS (max)TPS (mean)TPS (median)TPS (min)TPS (stddev)TTFT (max)TTFT (mean)TTFT (median)TTFT (min)TTFT (stddev)
05eb076 — 2026-05-20 03:32:574.17 (-0.31%)4.14 (-0.58%)4.13 (-0.62%)4.13 (-0.53%)0.01 (+28.64%)2.09 (-0.43%)2.08 (-0.28%)2.08 (+0.05%)2.06 (-0.43%)0.01 (-8.30%)
5503a95 — 2026-05-11 23:56:254.18 (n/a)4.17 (n/a)4.16 (n/a)4.15 (n/a)0.01 (n/a)2.10 (n/a)2.08 (n/a)2.08 (n/a)2.07 (n/a)0.01 (n/a)
Phoenix - Small

IRON

Tested on 2026_05_20_03_28_51 at commit 05eb076.

iron/operators/axpy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0]✅ 5/5330.120.04n/a
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]✅ 5/5549.880.03n/a
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]✅ 5/5334.560.04n/a
iron/operators/dequant
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32]✅ 5/5349.900.02n/a
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]✅ 5/5461.300.01n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]✅ 5/5392.720.01n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]✅ 5/5473.380.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]✅ 5/5547.340.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]✅ 5/5455.440.01n/a
iron/operators/elementwise_add
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5341.800.04n/a
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5437.280.03n/a
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5488.160.03n/a
iron/operators/elementwise_mul
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5263.600.05n/a
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5483.680.03n/a
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5388.340.04n/a
iron/operators/gelu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5387.720.02n/a
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5321.160.03n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5307.540.03n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5456.320.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5442.780.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5442.440.02n/a
iron/operators/gemm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1]✅ 5/5722.840.4017.14
test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1]✅ 5/5545.600.4217.72
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]✅ 5/584194.240.30204.08
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]✅ 5/525421.120.99675.84
test_gemm[M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1]✅ 5/54790.381.99523.09
test_gemm[M_64-K_512-N_256-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_16-k_64-n_64-trace_size_0-partition_N_4]✅ 5/55724.560.2212.02
iron/operators/gemv
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128]✅ 5/5n/a0.110.11
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a3.693.69
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]✅ 5/5n/a6.396.39
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]✅ 5/5n/a8.408.40
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a3.593.59
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a5.785.78
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a9.709.70
iron/operators/layer_norm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5306.200.03n/a
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5333.280.03n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5469.140.02n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5431.900.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5446.720.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5423.600.02n/a
iron/operators/mem_copy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048]✅ 5/5434.960.02n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]✅ 5/5320.280.03n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]✅ 5/5391.180.02n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]✅ 5/5514.280.02n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]✅ 5/5487.040.02n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]✅ 5/5513.600.02n/a
iron/operators/relu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5337.020.03n/a
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5481.780.02n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5545.920.02n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5389.700.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5379.380.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5464.520.02n/a
iron/operators/rms_norm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False]✅ 5/5356.960.03n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]✅ 5/5266.380.05n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]✅ 5/5493.980.02n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]✅ 5/5460.180.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]✅ 5/5409.420.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]✅ 5/5429.520.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]✅ 5/5410.040.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]✅ 5/5405.640.03n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]✅ 5/5352.060.03n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]✅ 5/5357.380.03n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]✅ 5/5480.800.02n/a
iron/operators/rope
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0]✅ 5/5294.180.34n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]✅ 5/5408.720.26n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]✅ 5/5374.500.31n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]✅ 5/5369.480.22n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]✅ 5/5443.160.17n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]✅ 5/5433.720.18n/a
iron/operators/sigmoid
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5336.480.03n/a
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5495.300.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5357.700.03n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5352.560.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5751.560.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5537.140.02n/a
iron/operators/silu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_silu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5324.920.03n/a
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5361.460.03n/a
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5371.900.02n/a
iron/operators/softmax
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024]✅ 5/5627.860.24n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]✅ 5/5508.200.29n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5527.340.26n/a
iron/operators/swiglu_decode
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584]✅ 5/514924.960.00n/a
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]✅ 5/519260.950.00n/a
iron/operators/swiglu_prefill
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False]✅ 5/524045.090.09n/a
iron/operators/tanh
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_tanh[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5264.020.03n/a
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5798.280.02n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5337.020.03n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5343.920.03n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5314.080.03n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5416.360.02n/a
iron/operators/transpose
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8]✅ 5/5513.561.10n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8]✅ 5/5454.261.22n/a

Trends:

IRON Trends

iron/operators/axpy

test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (+3.57%)0.04 (+13.19%)0.04 (+7.89%)0.02 (+23.91%)0.01 (-12.68%)514.50 (-19.31%)330.12 (-16.22%)280.00 (-7.32%)242.80 (-3.42%)114.43 (-32.69%)
5503a95 — 2026-05-11 23:50:480.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)637.60 (n/a)394.04 (n/a)302.10 (n/a)251.40 (n/a)170.01 (n/a)

test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (+1.51%)0.03 (-13.90%)0.04 (-22.13%)0.01 (-57.56%)0.02 (+21.16%)1398.10 (+135.65%)549.88 (+53.47%)318.40 (+28.44%)232.80 (-1.48%)491.17 (+194.97%)
5503a95 — 2026-05-11 23:50:480.05 (n/a)0.04 (n/a)0.05 (n/a)0.02 (n/a)0.02 (n/a)593.30 (n/a)358.30 (n/a)247.90 (n/a)236.30 (n/a)166.51 (n/a)

test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (+1.53%)0.04 (+17.57%)0.04 (+19.02%)0.03 (+45.85%)0.01 (-29.43%)436.40 (-31.44%)334.56 (-22.50%)295.20 (-15.99%)244.30 (-1.49%)82.95 (-54.38%)
5503a95 — 2026-05-11 23:50:480.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)636.50 (n/a)431.70 (n/a)351.40 (n/a)248.00 (n/a)181.83 (n/a)
iron/operators/dequant

test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+25.48%)0.02 (+10.04%)0.02 (+31.67%)0.01 (-3.69%)0.01 (+30.12%)499.30 (+3.83%)349.90 (-5.50%)313.30 (-24.07%)181.40 (-20.30%)137.97 (+12.71%)
5503a95 — 2026-05-11 23:50:480.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)480.90 (n/a)370.26 (n/a)412.60 (n/a)227.60 (n/a)122.40 (n/a)

test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+46.71%)0.01 (-2.03%)0.01 (-1.26%)0.01 (-21.62%)0.01 (+106.70%)623.00 (+27.59%)461.30 (+15.02%)461.00 (+1.27%)193.20 (-31.85%)167.94 (+69.69%)
5503a95 — 2026-05-11 23:50:480.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)488.30 (n/a)401.06 (n/a)455.20 (n/a)283.50 (n/a)98.97 (n/a)

test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.02 (-10.62%)0.01 (-6.34%)0.01 (-14.96%)0.01 (+12.19%)0.00 (-28.96%)482.50 (-10.86%)392.72 (+3.10%)433.50 (+17.61%)295.20 (+11.86%)82.42 (-28.97%)
5503a95 — 2026-05-11 23:50:480.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)541.30 (n/a)380.92 (n/a)368.60 (n/a)263.90 (n/a)116.04 (n/a)

test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+63.35%)0.01 (+16.39%)0.01 (-9.80%)0.01 (-1.12%)0.01 (+142.61%)618.20 (+1.13%)473.38 (-3.06%)523.90 (+10.85%)193.30 (-38.79%)165.30 (+38.02%)
5503a95 — 2026-05-11 23:50:480.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)611.30 (n/a)488.34 (n/a)472.60 (n/a)315.80 (n/a)119.77 (n/a)

test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.02 (-0.41%)0.01 (-28.82%)0.01 (-42.74%)0.00 (-53.95%)0.01 (+14.23%)1061.60 (+117.14%)547.34 (+59.37%)486.50 (+74.69%)245.80 (+0.41%)305.62 (+156.90%)
5503a95 — 2026-05-11 23:50:480.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)488.90 (n/a)343.44 (n/a)278.50 (n/a)244.80 (n/a)118.97 (n/a)

test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.02 (-16.01%)0.01 (-7.53%)0.01 (+0.12%)0.01 (+16.35%)0.00 (-33.16%)512.90 (-14.06%)455.44 (+2.62%)495.30 (-0.12%)285.40 (+19.07%)96.52 (-30.51%)
5503a95 — 2026-05-11 23:50:480.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)596.80 (n/a)443.82 (n/a)495.90 (n/a)239.70 (n/a)138.91 (n/a)
iron/operators/elementwise_add

test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)506.70 (n/a)341.80 (n/a)308.40 (n/a)264.70 (n/a)99.50 (n/a)

test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)595.80 (n/a)437.28 (n/a)450.60 (n/a)268.00 (n/a)118.57 (n/a)

test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)573.10 (n/a)488.16 (n/a)519.70 (n/a)383.60 (n/a)92.53 (n/a)
iron/operators/elementwise_mul

test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (n/a)0.05 (n/a)0.05 (n/a)0.04 (n/a)0.00 (n/a)298.20 (n/a)263.60 (n/a)265.80 (n/a)228.90 (n/a)26.89 (n/a)

test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)574.10 (n/a)483.68 (n/a)549.30 (n/a)248.90 (n/a)135.18 (n/a)

test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.08 (n/a)0.04 (n/a)0.03 (n/a)0.02 (n/a)0.03 (n/a)551.50 (n/a)388.34 (n/a)480.00 (n/a)151.60 (n/a)176.98 (n/a)
iron/operators/gelu

test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)581.90 (n/a)387.72 (n/a)295.20 (n/a)264.30 (n/a)144.26 (n/a)

test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)431.90 (n/a)321.16 (n/a)287.60 (n/a)236.50 (n/a)82.53 (n/a)

test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)469.90 (n/a)307.54 (n/a)285.80 (n/a)211.70 (n/a)98.29 (n/a)

test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)626.40 (n/a)456.32 (n/a)485.80 (n/a)305.00 (n/a)136.44 (n/a)

test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)664.00 (n/a)442.78 (n/a)437.40 (n/a)196.70 (n/a)210.38 (n/a)

test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)625.00 (n/a)442.44 (n/a)451.50 (n/a)319.10 (n/a)119.17 (n/a)
iron/operators/gemm

test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_False-c_col_maj_False-m_48-k_96-n_16-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:580.54 (-16.08%)0.40 (-3.72%)0.47 (+20.79%)0.13 (-31.41%)0.16 (-6.27%)1707.10 (+45.81%)722.84 (+14.93%)468.80 (-17.22%)413.40 (+19.17%)552.89 (+70.83%)22.83 (-16.08%)17.14 (-3.72%)20.13 (+20.79%)5.53 (-31.41%)6.84 (-6.27%)
5503a95 — 2026-05-11 23:50:480.64 (n/a)0.42 (n/a)0.39 (n/a)0.19 (n/a)0.17 (n/a)1170.80 (n/a)628.94 (n/a)566.30 (n/a)346.90 (n/a)323.64 (n/a)27.20 (n/a)17.80 (n/a)16.67 (n/a)8.06 (n/a)7.30 (n/a)

test_gemm[M_192-K_384-N_64-num_aie_columns_4-b_col_maj_True-c_col_maj_True-m_48-k_96-n_16-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:580.54 (+0.41%)0.42 (+10.98%)0.40 (+8.11%)0.35 (+173.48%)0.08 (-53.43%)625.20 (-63.44%)545.60 (-28.85%)552.30 (-7.50%)409.50 (-0.41%)88.44 (-83.59%)23.05 (+0.41%)17.72 (+10.98%)17.09 (+8.11%)15.09 (+173.48%)3.25 (-53.43%)
5503a95 — 2026-05-11 23:50:480.54 (n/a)0.37 (n/a)0.37 (n/a)0.13 (n/a)0.16 (n/a)1709.90 (n/a)766.82 (n/a)597.10 (n/a)411.20 (n/a)538.94 (n/a)22.95 (n/a)15.96 (n/a)15.81 (n/a)5.52 (n/a)6.98 (n/a)

test_gemm[M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:580.30 (-2.11%)0.30 (-2.33%)0.30 (-2.15%)0.29 (-3.36%)0.00 (+33.11%)86110.00 (+3.48%)84194.24 (+2.39%)83859.40 (+2.20%)82868.40 (+2.15%)1189.62 (+40.96%)207.32 (-2.11%)204.08 (-2.33%)204.87 (-2.15%)199.51 (-3.36%)2.86 (+33.11%)
5503a95 — 2026-05-11 23:50:480.31 (n/a)0.31 (n/a)0.31 (n/a)0.30 (n/a)0.00 (n/a)83216.60 (n/a)82225.06 (n/a)82057.10 (n/a)81121.80 (n/a)843.94 (n/a)211.78 (n/a)208.95 (n/a)209.36 (n/a)206.45 (n/a)2.15 (n/a)

test_gemm[M_2048-K_2048-N_2048-num_aie_columns_2-b_col_maj_True-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:581.00 (-0.72%)0.99 (-0.01%)0.99 (-0.17%)0.98 (+0.91%)0.01 (-47.26%)25630.40 (-0.90%)25421.12 (+0.00%)25430.20 (+0.17%)25198.00 (+0.72%)180.74 (-47.36%)681.79 (-0.72%)675.84 (-0.01%)675.57 (-0.17%)670.29 (+0.91%)4.81 (-47.26%)
5503a95 — 2026-05-11 23:50:481.01 (n/a)0.99 (n/a)0.99 (n/a)0.97 (n/a)0.01 (n/a)25863.90 (n/a)25420.24 (n/a)25386.80 (n/a)25017.00 (n/a)343.35 (n/a)686.73 (n/a)675.93 (n/a)676.72 (n/a)664.24 (n/a)9.12 (n/a)

test_gemm[M_384-K_1536-N_1792-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_32-k_48-n_64-trace_size_0-partition_N_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:583.82 (-3.84%)1.99 (-27.33%)1.57 (-25.47%)1.15 (-36.99%)1.07 (-2.29%)6984.60 (+58.69%)4790.38 (+44.82%)5125.10 (+34.17%)2111.70 (+3.99%)1851.21 (+58.31%)1001.07 (-3.84%)523.09 (-27.33%)412.47 (-25.47%)302.66 (-36.99%)280.05 (-2.29%)
5503a95 — 2026-05-11 23:50:483.97 (n/a)2.74 (n/a)2.11 (n/a)1.83 (n/a)1.09 (n/a)4401.30 (n/a)3307.84 (n/a)3819.90 (n/a)2030.60 (n/a)1169.34 (n/a)1041.02 (n/a)719.80 (n/a)553.40 (n/a)480.30 (n/a)286.60 (n/a)

test_gemm[M_64-K_512-N_256-num_aie_columns_4-b_col_maj_True-c_col_maj_False-m_16-k_64-n_64-trace_size_0-partition_N_4]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:580.27 (+3.34%)0.22 (+5.56%)0.22 (-5.00%)0.18 (+16.85%)0.04 (-13.60%)7058.00 (-14.42%)5724.56 (-6.76%)5611.10 (+5.27%)4652.30 (-3.23%)1025.36 (-29.50%)14.42 (+3.34%)12.02 (+5.56%)11.96 (-5.00%)9.51 (+16.85%)2.11 (-13.60%)
5503a95 — 2026-05-11 23:50:480.26 (n/a)0.21 (n/a)0.23 (n/a)0.15 (n/a)0.05 (n/a)8247.30 (n/a)6139.54 (n/a)5330.30 (n/a)4807.80 (n/a)1454.42 (n/a)13.96 (n/a)11.39 (n/a)12.59 (n/a)8.14 (n/a)2.45 (n/a)
iron/operators/gemv

test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:580.13 (+8.59%)0.11 (+65.59%)0.12 (+88.26%)0.09 (+170.21%)0.02 (-44.98%)0.13 (+8.59%)0.11 (+65.59%)0.12 (+88.26%)0.08 (+170.21%)0.02 (-44.98%)
5503a95 — 2026-05-11 23:50:480.12 (n/a)0.07 (n/a)0.06 (n/a)0.03 (n/a)0.03 (n/a)0.12 (n/a)0.07 (n/a)0.06 (n/a)0.03 (n/a)0.03 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:583.87 (+0.37%)3.69 (+2.64%)3.72 (+4.01%)3.32 (+0.11%)0.22 (+11.18%)3.87 (+0.37%)3.69 (+2.64%)3.72 (+4.01%)3.32 (+0.11%)0.22 (+11.18%)
5503a95 — 2026-05-11 23:50:483.86 (n/a)3.59 (n/a)3.58 (n/a)3.32 (n/a)0.20 (n/a)3.85 (n/a)3.59 (n/a)3.57 (n/a)3.32 (n/a)0.19 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:587.02 (-6.74%)6.39 (-10.42%)6.78 (-4.76%)5.61 (-14.66%)0.69 (+84.29%)7.02 (-6.74%)6.39 (-10.42%)6.78 (-4.76%)5.61 (-14.66%)0.69 (+84.29%)
5503a95 — 2026-05-11 23:50:487.53 (n/a)7.13 (n/a)7.12 (n/a)6.57 (n/a)0.37 (n/a)7.53 (n/a)7.13 (n/a)7.12 (n/a)6.57 (n/a)0.37 (n/a)

test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:589.34 (-32.62%)8.40 (-19.59%)8.40 (-14.41%)7.74 (-2.22%)0.60 (-76.96%)9.33 (-32.62%)8.40 (-19.59%)8.39 (-14.41%)7.74 (-2.22%)0.60 (-76.96%)
5503a95 — 2026-05-11 23:50:4813.86 (n/a)10.45 (n/a)9.81 (n/a)7.92 (n/a)2.58 (n/a)13.85 (n/a)10.44 (n/a)9.80 (n/a)7.91 (n/a)2.58 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:583.73 (-1.45%)3.59 (+1.87%)3.71 (-0.73%)3.37 (+9.80%)0.18 (-45.89%)3.73 (-1.45%)3.59 (+1.87%)3.70 (-0.73%)3.37 (+9.80%)0.18 (-45.89%)
5503a95 — 2026-05-11 23:50:483.79 (n/a)3.52 (n/a)3.73 (n/a)3.07 (n/a)0.33 (n/a)3.79 (n/a)3.52 (n/a)3.73 (n/a)3.07 (n/a)0.33 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:586.44 (-10.88%)5.78 (-13.17%)5.78 (-15.50%)5.00 (-11.76%)0.52 (-11.56%)6.43 (-10.88%)5.78 (-13.17%)5.78 (-15.50%)4.99 (-11.76%)0.52 (-11.56%)
5503a95 — 2026-05-11 23:50:487.22 (n/a)6.66 (n/a)6.85 (n/a)5.66 (n/a)0.59 (n/a)7.22 (n/a)6.65 (n/a)6.84 (n/a)5.66 (n/a)0.59 (n/a)

test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
05eb076 — 2026-05-20 03:25:5813.96 (-1.12%)9.70 (-9.59%)8.53 (-21.56%)7.85 (+11.40%)2.47 (-2.88%)13.95 (-1.12%)9.70 (-9.59%)8.53 (-21.56%)7.84 (+11.40%)2.47 (-2.88%)
5503a95 — 2026-05-11 23:50:4814.12 (n/a)10.73 (n/a)10.88 (n/a)7.04 (n/a)2.54 (n/a)14.11 (n/a)10.73 (n/a)10.87 (n/a)7.04 (n/a)2.54 (n/a)
iron/operators/layer_norm

test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)448.90 (n/a)306.20 (n/a)280.20 (n/a)238.50 (n/a)83.63 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)563.80 (n/a)333.28 (n/a)285.80 (n/a)247.10 (n/a)130.48 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)612.40 (n/a)469.14 (n/a)553.50 (n/a)265.00 (n/a)162.65 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)657.50 (n/a)431.90 (n/a)508.90 (n/a)235.70 (n/a)184.13 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)512.20 (n/a)446.72 (n/a)479.40 (n/a)275.40 (n/a)97.96 (n/a)

test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)650.80 (n/a)423.60 (n/a)349.20 (n/a)248.50 (n/a)178.94 (n/a)
iron/operators/mem_copy

test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (-16.15%)0.02 (-5.07%)0.02 (+18.80%)0.01 (-22.02%)0.01 (-18.33%)684.20 (+28.22%)434.96 (+4.49%)424.20 (-15.83%)284.40 (+19.25%)162.70 (+16.22%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)533.60 (n/a)416.26 (n/a)504.00 (n/a)238.50 (n/a)139.99 (n/a)

test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+11.80%)0.03 (+20.63%)0.03 (+9.00%)0.02 (+51.90%)0.00 (-41.15%)404.90 (-34.16%)320.28 (-24.36%)298.40 (-8.27%)257.70 (-10.55%)57.58 (-65.51%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)615.00 (n/a)423.44 (n/a)325.30 (n/a)288.10 (n/a)166.95 (n/a)

test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+0.99%)0.02 (+9.07%)0.02 (+26.63%)0.01 (+3.82%)0.01 (+14.91%)564.90 (-3.68%)391.18 (-6.45%)355.90 (-21.02%)263.40 (-0.94%)139.28 (+8.88%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)586.50 (n/a)418.16 (n/a)450.60 (n/a)265.90 (n/a)127.92 (n/a)

test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (+25.34%)0.02 (-4.46%)0.02 (-8.16%)0.01 (-33.86%)0.01 (+84.46%)862.80 (+51.21%)514.28 (+19.95%)438.30 (+8.87%)233.70 (-20.21%)239.99 (+119.16%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)570.60 (n/a)428.74 (n/a)402.60 (n/a)292.90 (n/a)109.50 (n/a)

test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.02 (-47.10%)0.02 (-40.01%)0.02 (-42.49%)0.02 (-18.24%)0.00 (-70.69%)524.50 (+22.32%)487.04 (+53.83%)519.60 (+73.90%)364.80 (+89.02%)68.88 (-36.56%)
5503a95 — 2026-05-11 23:50:480.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)428.80 (n/a)316.60 (n/a)298.80 (n/a)193.00 (n/a)108.58 (n/a)

test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.02 (-39.22%)0.02 (-12.61%)0.02 (+22.19%)0.01 (-5.90%)0.00 (-57.25%)663.40 (+6.26%)513.60 (+5.98%)460.20 (-18.17%)408.80 (+64.51%)117.48 (-24.82%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)624.30 (n/a)484.62 (n/a)562.40 (n/a)248.50 (n/a)156.26 (n/a)
iron/operators/rms_norm

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (-2.10%)0.03 (+5.60%)0.03 (+21.93%)0.01 (-6.48%)0.01 (+5.68%)590.10 (+6.94%)356.96 (-3.44%)293.00 (-17.97%)242.90 (+2.14%)142.49 (+16.82%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)551.80 (n/a)369.66 (n/a)357.20 (n/a)237.80 (n/a)121.97 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.05 (+46.44%)0.05 (+76.42%)0.05 (+105.22%)0.04 (+102.66%)0.00 (-37.84%)290.40 (-50.66%)266.38 (-45.58%)272.00 (-51.28%)233.80 (-31.72%)22.93 (-79.33%)
5503a95 — 2026-05-11 23:50:480.04 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)588.60 (n/a)489.52 (n/a)558.30 (n/a)342.40 (n/a)110.97 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (-7.04%)0.02 (+3.38%)0.02 (+0.49%)0.01 (-15.77%)0.01 (-7.00%)746.40 (+18.72%)493.98 (-2.27%)518.70 (-0.48%)275.50 (+7.58%)180.63 (+22.09%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)628.70 (n/a)505.44 (n/a)521.20 (n/a)256.10 (n/a)147.95 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (-49.10%)0.02 (-31.98%)0.02 (-15.83%)0.02 (+2.11%)0.01 (-65.15%)569.80 (-2.08%)460.18 (+21.36%)521.10 (+18.81%)278.20 (+96.47%)123.44 (-31.71%)
5503a95 — 2026-05-11 23:50:480.07 (n/a)0.04 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)581.90 (n/a)379.18 (n/a)438.60 (n/a)141.60 (n/a)180.76 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+17.07%)0.02 (+28.69%)0.02 (+12.03%)0.01 (+326.24%)0.01 (-7.58%)573.40 (-76.54%)409.42 (-50.24%)441.40 (-10.74%)238.60 (-14.57%)149.95 (-83.58%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)0.01 (n/a)2444.20 (n/a)822.84 (n/a)494.50 (n/a)279.30 (n/a)913.16 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (-37.44%)0.03 (-20.58%)0.02 (-17.77%)0.02 (-4.43%)0.01 (-52.52%)544.20 (+4.63%)429.52 (+15.60%)443.30 (+21.62%)271.00 (+59.88%)112.83 (-15.39%)
5503a95 — 2026-05-11 23:50:480.06 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)520.10 (n/a)371.56 (n/a)364.50 (n/a)169.50 (n/a)133.35 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (+1.46%)0.02 (+44.77%)0.02 (+54.90%)0.01 (+233.48%)0.01 (-27.24%)593.50 (-70.02%)410.04 (-50.51%)412.40 (-35.44%)264.00 (-1.46%)134.74 (-79.94%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)0.01 (n/a)1979.40 (n/a)828.60 (n/a)638.80 (n/a)267.90 (n/a)671.83 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (-14.44%)0.03 (-0.00%)0.03 (+17.79%)0.01 (-11.58%)0.01 (-14.83%)676.50 (+13.09%)405.64 (-0.69%)298.20 (-15.09%)228.10 (+16.85%)189.98 (+9.33%)
5503a95 — 2026-05-11 23:50:480.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)598.20 (n/a)408.44 (n/a)351.20 (n/a)195.20 (n/a)173.77 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (-11.48%)0.03 (+23.34%)0.03 (+85.78%)0.01 (-4.53%)0.01 (-9.89%)593.90 (+4.76%)352.06 (-19.29%)272.00 (-46.18%)241.20 (+12.92%)151.52 (+6.40%)
5503a95 — 2026-05-11 23:50:480.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)566.90 (n/a)436.22 (n/a)505.40 (n/a)213.60 (n/a)142.41 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.04 (+10.58%)0.03 (+38.42%)0.03 (+99.06%)0.02 (+20.70%)0.01 (+14.96%)520.80 (-17.15%)357.38 (-27.39%)278.20 (-49.77%)244.00 (-9.60%)129.54 (-9.52%)
5503a95 — 2026-05-11 23:50:480.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)628.60 (n/a)492.22 (n/a)553.80 (n/a)269.90 (n/a)143.18 (n/a)

test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.03 (-22.41%)0.02 (-4.49%)0.02 (-4.06%)0.01 (+61.69%)0.01 (-42.19%)590.20 (-38.16%)480.80 (-9.58%)501.10 (+4.22%)281.30 (+28.92%)125.08 (-53.03%)
5503a95 — 2026-05-11 23:50:480.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)954.40 (n/a)531.76 (n/a)480.80 (n/a)218.20 (n/a)266.30 (n/a)
iron/operators/rope

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.38 (+9.19%)0.34 (+34.14%)0.33 (+34.89%)0.31 (+115.67%)0.03 (-67.33%)315.40 (-53.64%)294.18 (-32.83%)295.70 (-25.87%)256.60 (-8.42%)23.52 (-85.89%)
5503a95 — 2026-05-11 23:50:480.35 (n/a)0.25 (n/a)0.25 (n/a)0.14 (n/a)0.09 (n/a)680.30 (n/a)437.96 (n/a)398.90 (n/a)280.20 (n/a)166.68 (n/a)

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.40 (+71.44%)0.26 (+32.30%)0.23 (+18.45%)0.18 (+10.83%)0.09 (+237.82%)535.00 (-9.77%)408.72 (-19.05%)427.40 (-15.58%)243.10 (-41.67%)120.28 (+79.19%)
5503a95 — 2026-05-11 23:50:480.24 (n/a)0.20 (n/a)0.19 (n/a)0.17 (n/a)0.03 (n/a)592.90 (n/a)504.90 (n/a)506.30 (n/a)416.80 (n/a)67.12 (n/a)

test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.44 (+74.01%)0.31 (+57.52%)0.37 (+87.71%)0.16 (-6.88%)0.13 (+318.22%)614.00 (+7.38%)374.50 (-25.50%)266.70 (-46.73%)225.80 (-42.53%)183.98 (+156.00%)
5503a95 — 2026-05-11 23:50:480.25 (n/a)0.20 (n/a)0.20 (n/a)0.17 (n/a)0.03 (n/a)571.80 (n/a)502.68 (n/a)500.70 (n/a)392.90 (n/a)71.87 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.30 (+28.95%)0.22 (+54.69%)0.24 (+57.74%)0.11 (+198.02%)0.07 (+2.63%)655.10 (-66.45%)369.48 (-50.46%)309.70 (-36.61%)247.90 (-22.46%)164.67 (-75.73%)
5503a95 — 2026-05-11 23:50:480.23 (n/a)0.14 (n/a)0.15 (n/a)0.04 (n/a)0.07 (n/a)1952.40 (n/a)745.78 (n/a)488.60 (n/a)319.70 (n/a)678.37 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.24 (-31.22%)0.17 (-27.02%)0.17 (-3.29%)0.13 (-17.83%)0.04 (-52.89%)553.70 (+21.69%)443.16 (+27.99%)437.00 (+3.38%)304.30 (+45.39%)104.80 (-13.15%)
5503a95 — 2026-05-11 23:50:480.35 (n/a)0.24 (n/a)0.17 (n/a)0.16 (n/a)0.10 (n/a)455.00 (n/a)346.24 (n/a)422.70 (n/a)209.30 (n/a)120.66 (n/a)

test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.24 (-10.20%)0.18 (-8.24%)0.17 (-18.61%)0.13 (+10.42%)0.05 (-32.39%)553.40 (-9.43%)433.72 (+2.04%)438.10 (+22.89%)303.00 (+11.36%)113.31 (-33.12%)
5503a95 — 2026-05-11 23:50:480.27 (n/a)0.20 (n/a)0.21 (n/a)0.12 (n/a)0.07 (n/a)611.00 (n/a)425.06 (n/a)356.50 (n/a)272.10 (n/a)169.41 (n/a)
iron/operators/softmax

test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.35 (-22.60%)0.24 (-21.36%)0.24 (-16.23%)0.11 (-33.94%)0.08 (-22.06%)1159.00 (+51.38%)627.86 (+30.79%)535.40 (+19.38%)378.30 (+29.20%)304.92 (+65.70%)
5503a95 — 2026-05-11 23:50:480.45 (n/a)0.30 (n/a)0.29 (n/a)0.17 (n/a)0.11 (n/a)765.60 (n/a)480.06 (n/a)448.50 (n/a)292.80 (n/a)184.02 (n/a)

test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.48 (+1.90%)0.29 (-12.74%)0.26 (-25.32%)0.18 (-0.50%)0.11 (+5.77%)714.80 (+0.51%)508.20 (+15.10%)502.60 (+33.92%)275.80 (-1.85%)168.28 (-0.74%)
5503a95 — 2026-05-11 23:50:480.47 (n/a)0.33 (n/a)0.35 (n/a)0.18 (n/a)0.11 (n/a)711.20 (n/a)441.52 (n/a)375.30 (n/a)281.00 (n/a)169.54 (n/a)

test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.39 (-44.85%)0.26 (-40.24%)0.23 (-43.97%)0.20 (-26.72%)0.08 (-51.83%)655.80 (+36.48%)527.34 (+61.43%)572.80 (+78.50%)337.20 (+81.29%)131.26 (+22.36%)
5503a95 — 2026-05-11 23:50:480.70 (n/a)0.44 (n/a)0.41 (n/a)0.27 (n/a)0.16 (n/a)480.50 (n/a)326.66 (n/a)320.90 (n/a)186.00 (n/a)107.27 (n/a)
iron/operators/swiglu_decode

test_swiglu_decode[embedding_dim_1024-hidden_dim_3584]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.00 (+0.00%)0.00 (+11.76%)0.00 (-33.33%)0.00 (+0.00%)0.00 (+20.08%)21270.49 (-6.85%)14924.96 (-2.86%)20067.66 (+29.68%)5800.41 (-7.20%)7826.43 (+30.05%)
5503a95 — 2026-05-11 23:50:480.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)22834.34 (n/a)15364.62 (n/a)15474.57 (n/a)6250.44 (n/a)6018.20 (n/a)

test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.00 (-16.67%)0.00 (-15.38%)0.00 (-20.00%)0.00 (+0.00%)0.00 (-34.53%)22628.36 (+17.61%)19260.95 (+19.50%)19204.29 (+14.39%)15055.65 (+11.67%)3162.34 (+37.19%)
5503a95 — 2026-05-11 23:50:480.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)19240.68 (n/a)16117.52 (n/a)16788.38 (n/a)13481.94 (n/a)2305.00 (n/a)
iron/operators/swiglu_prefill

test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:580.13 (-4.22%)0.09 (-8.17%)0.08 (-7.85%)0.07 (-13.39%)0.02 (+9.44%)28704.82 (+15.48%)24045.09 (+10.27%)25177.84 (+8.56%)16218.60 (+4.41%)4872.84 (+32.63%)
5503a95 — 2026-05-11 23:50:480.14 (n/a)0.10 (n/a)0.09 (n/a)0.08 (n/a)0.02 (n/a)24856.64 (n/a)21805.69 (n/a)23192.07 (n/a)15534.00 (n/a)3674.02 (n/a)
iron/operators/transpose

test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:581.72 (+6.19%)1.10 (+0.86%)0.95 (-14.21%)0.80 (+393.26%)0.37 (-38.19%)659.20 (-79.73%)513.56 (-47.86%)551.00 (+16.56%)305.60 (-5.82%)135.63 (-89.33%)
5503a95 — 2026-05-11 23:50:481.62 (n/a)1.09 (n/a)1.11 (n/a)0.16 (n/a)0.59 (n/a)3251.70 (n/a)984.96 (n/a)472.70 (n/a)324.50 (n/a)1270.69 (n/a)

test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
05eb076 — 2026-05-20 03:25:581.67 (+14.07%)1.22 (+13.38%)1.09 (+13.39%)0.86 (-1.11%)0.32 (+33.01%)609.20 (+1.13%)454.26 (-10.16%)481.50 (-11.81%)314.30 (-12.33%)113.63 (+18.31%)
5503a95 — 2026-05-11 23:50:481.46 (n/a)1.07 (n/a)0.96 (n/a)0.87 (n/a)0.24 (n/a)602.40 (n/a)505.64 (n/a)546.00 (n/a)358.50 (n/a)96.04 (n/a)
Phoenix - Examples

IRON

Tested on 2026_05_20_03_30_48 at commit 05eb076.

Trends:

IRON Trends

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@andrej andrej Awaiting requested review from andrej andrej is a code owner

@jgmelber jgmelber Awaiting requested review from jgmelber jgmelber is a code owner

@hunhoffe hunhoffe Awaiting requested review from hunhoffe hunhoffe is a code owner

At least 1 approving review is required to merge this pull request.

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@x4v13r64 @andrej @jgmelber