Skip to content

compilation: honor AIECC_PATH and AIECC_JOBS env overrides#134

Open
atassis wants to merge 1 commit into
amd:develfrom
atassis:feat/aiecc-path-jobs-env
Open

compilation: honor AIECC_PATH and AIECC_JOBS env overrides#134
atassis wants to merge 1 commit into
amd:develfrom
atassis:feat/aiecc-path-jobs-env

Conversation

@atassis

@atassis atassis commented Jul 10, 2026

Copy link
Copy Markdown
Contributor

Two opt-in, backward-compatible build-ergonomics knobs on the aiecc compilation rules. Both default to the current behavior, so nothing changes unless the env var is set.

  • AIECC_PATH - point the compilation rules at a locally-built aiecc (for example a compiler under active development) instead of the one under mlir_aie_dir/bin, without having to overwrite the installed binary. Default: the installed aiecc.
  • AIECC_JOBS - pass -j<N> to aiecc instead of the hardcoded -j1, so multi-core hosts can compile the AIE cores in parallel. Default 1 = unchanged.

Both are read via os.environ.get(...) in iron/common/compilation/base.py (the aiecc path resolution and the two Aiecc*CompilationRule command builders). No new dependencies.

@github-actions

github-actions Bot commented Jul 13, 2026

Copy link
Copy Markdown
Contributor

CI Test Results

6ff8ed1 (2026_07_14_16_39_45)

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 Phoenix Status Phoenix Latency (mean)
test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0] - - 331.46
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0] - - 477.36
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0] - - 464.74
iron/operators/dequant
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32] - - 439.40
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32] - - 489.18
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32] - - 728.80
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32] - - 575.30
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32] - - 464.72
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32] - - 494.44
iron/operators/elementwise_add
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048] - - 384.02
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024] - - 343.82
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512] - - 592.22
iron/operators/elementwise_mul
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048] - - 353.94
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024] - - 377.86
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512] - - 456.48
iron/operators/gelu
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] - - 362.18
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 756.18
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 413.40
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 444.86
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 532.68
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 416.32
iron/operators/gemm
Test Krackan Status Krackan Phoenix Status Phoenix Latency (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] - - 531.06
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] - - 540.22
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] - - 82502.90
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] - - 24866.80
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] - - 4348.00
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] - - 5625.26
iron/operators/gemv
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128] - - 0.10
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048] - - 3.69
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024] - - 6.11
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512] - - 10.84
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024] - - 3.70
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024] - - 6.14
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024] - - 9.39
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2] - - 2.33
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2] - - 0.31
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4] - - 0.66
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8] - - 1.54
iron/operators/layer_norm
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] - - 379.24
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 302.86
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 324.16
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 369.20
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 412.00
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 410.94
iron/operators/leaky_relu
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.01] - - 341.76
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1] - - 339.56
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25] - - 323.92
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01] - - 607.48
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01] - - 343.24
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01] - - 356.76
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01] - - 358.86
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01] - - 412.24
iron/operators/mem_copy
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048] - - 495.30
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024] - - 413.08
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024] - - 329.14
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512] - - 582.78
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512] - - 411.96
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256] - - 465.26
iron/operators/relu
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] - - 311.98
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 489.14
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 358.44
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 495.66
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 482.12
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 506.42
iron/operators/rms_norm
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False] - - 437.28
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True] - - 399.72
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False] - - 434.02
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True] - - 392.72
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False] - - 348.22
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True] - - 385.98
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False] - - 382.02
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True] - - 528.56
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False] - - 422.20
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True] - - 507.84
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False] - - 408.94
iron/operators/rope
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0] - - 449.36
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0] - - 481.08
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0] - - 483.56
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0] - - 693.76
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0] - - 348.64
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0] - - 500.70
iron/operators/sigmoid
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] - - 427.64
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 492.60
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 375.18
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 453.80
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 405.48
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 465.62
iron/operators/silu
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_silu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] - - 402.60
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 408.88
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 433.74
iron/operators/softmax
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024] - - 514.26
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048] - - 555.02
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512] - - 540.42
iron/operators/swiglu_decode
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584] - - 16940.51
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048] - - 15616.74
iron/operators/swiglu_prefill
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False] - - 24431.98
iron/operators/tanh
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_tanh[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] - - 324.52
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 379.54
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 379.90
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 459.94
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 464.34
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 409.12
iron/operators/transpose
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_1] - - 458.84
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2] - - 501.64
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1] - - 524.92
Krackan - Examples

IRON

Tested on 2026_07_14_16_39_45 at commit 6ff8ed1.

iron/applications/llama_3.2_1b
TestChecksTTFT (mean)TPS (mean)
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_1]✅ 5/52.13n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_40]✅ 5/52.167.76
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_1]✅ 5/52.08n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_40]✅ 5/52.087.53

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)
6ff8ed1 — 2026-07-14 16:34:442.14 (+0.28%)2.13 (+0.08%)2.13 (+0.28%)2.11 (+0.19%)0.01 (+13.43%)
e216c2f — 2026-07-13 21:54:522.13 (n/a)2.12 (n/a)2.12 (n/a)2.11 (n/a)0.01 (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)
6ff8ed1 — 2026-07-14 16:34:447.82 (+0.85%)7.76 (+0.40%)7.75 (+0.12%)7.72 (+0.25%)0.04 (+73.26%)2.30 (+0.66%)2.16 (-0.41%)2.13 (-0.51%)2.11 (-0.75%)0.08 (+20.32%)
e216c2f — 2026-07-13 21:54:527.76 (n/a)7.73 (n/a)7.74 (n/a)7.70 (n/a)0.02 (n/a)2.28 (n/a)2.17 (n/a)2.14 (n/a)2.13 (n/a)0.07 (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)
6ff8ed1 — 2026-07-14 16:34:442.10 (-0.52%)2.08 (-0.98%)2.07 (-1.33%)2.07 (-1.38%)0.02 (+117.78%)
e216c2f — 2026-07-13 21:54:522.11 (n/a)2.10 (n/a)2.10 (n/a)2.10 (n/a)0.01 (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)
6ff8ed1 — 2026-07-14 16:34:447.55 (+0.23%)7.53 (+0.12%)7.53 (+0.19%)7.50 (+0.13%)0.03 (+43.03%)2.11 (-0.05%)2.08 (-0.73%)2.07 (-1.19%)2.06 (-0.91%)0.02 (+72.88%)
e216c2f — 2026-07-13 21:54:527.53 (n/a)7.52 (n/a)7.52 (n/a)7.49 (n/a)0.02 (n/a)2.12 (n/a)2.10 (n/a)2.10 (n/a)2.08 (n/a)0.01 (n/a)
Phoenix - Small

IRON

Tested on 2026_07_14_16_28_27 at commit 6ff8ed1.

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]✅ 5/5331.460.04n/a
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]✅ 5/5477.360.03n/a
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]✅ 5/5464.740.03n/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]✅ 5/5439.400.02n/a
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]✅ 5/5489.180.01n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]✅ 5/5728.800.01n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]✅ 5/5575.300.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]✅ 5/5464.720.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]✅ 5/5494.440.01n/a
iron/operators/elementwise_add
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5384.020.03n/a
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5343.820.04n/a
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5592.220.03n/a
iron/operators/elementwise_mul
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5353.940.04n/a
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5377.860.04n/a
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5456.480.03n/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]✅ 5/5362.180.03n/a
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5756.180.02n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5413.400.02n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5444.860.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5532.680.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5416.320.02n/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]✅ 5/5531.060.4217.82
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]✅ 5/5540.220.4117.67
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]✅ 5/582502.900.31208.28
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]✅ 5/524866.801.01691.01
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]✅ 5/54348.002.24586.13
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]✅ 5/55625.260.2312.14
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/a0.100.10
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a3.693.68
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]✅ 5/5n/a6.116.11
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]✅ 5/5n/a10.8410.83
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a3.703.70
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a6.146.13
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a9.399.38
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2]✅ 5/5n/a2.332.33
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2]✅ 5/5n/a0.310.31
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4]✅ 5/5n/a0.660.65
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8]✅ 5/5n/a1.541.51
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]✅ 5/5379.240.03n/a
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5302.860.03n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5324.160.03n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5369.200.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5412.000.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5410.940.02n/a
iron/operators/leaky_relu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.01]✅ 5/5341.760.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1]✅ 5/5339.560.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25]✅ 5/5323.920.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01]✅ 5/5607.480.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01]✅ 5/5343.240.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01]✅ 5/5356.760.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01]✅ 5/5358.860.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01]✅ 5/5412.240.02n/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]✅ 5/5495.300.02n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]✅ 5/5413.080.02n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]✅ 5/5329.140.03n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]✅ 5/5582.780.02n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]✅ 5/5411.960.02n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]✅ 5/5465.260.02n/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]✅ 5/5311.980.03n/a
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5489.140.02n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5358.440.03n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5495.660.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5482.120.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5506.420.02n/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]✅ 5/5437.280.02n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]✅ 5/5399.720.03n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]✅ 5/5434.020.02n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]✅ 5/5392.720.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]✅ 5/5348.220.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]✅ 5/5385.980.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]✅ 5/5382.020.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]✅ 5/5528.560.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]✅ 5/5422.200.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]✅ 5/5507.840.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]✅ 5/5408.940.02n/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]✅ 5/5449.360.24n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]✅ 5/5481.080.26n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]✅ 5/5483.560.22n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]✅ 5/5693.760.18n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]✅ 5/5348.640.22n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]✅ 5/5500.700.17n/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]✅ 5/5427.640.02n/a
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5492.600.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5375.180.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5453.800.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5405.480.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5465.620.02n/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]✅ 5/5402.600.02n/a
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5408.880.02n/a
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5433.740.02n/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]✅ 5/5514.260.29n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]✅ 5/5555.020.26n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5540.420.26n/a
iron/operators/swiglu_decode
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584]✅ 5/516940.510.00n/a
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]✅ 5/515616.740.00n/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]✅ 5/524431.980.09n/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]✅ 5/5324.520.03n/a
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5379.540.02n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5379.900.03n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5459.940.02n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5464.340.02n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5409.120.02n/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-num_batches_1]✅ 5/5458.841.31n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2]✅ 5/5501.642.27n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1]✅ 5/5524.921.07n/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)
6ff8ed1 — 2026-07-14 16:24:460.05 (+3.57%)0.04 (+25.68%)0.04 (+53.34%)0.03 (+48.59%)0.01 (-46.71%)407.20 (-32.71%)331.46 (-26.79%)320.10 (-34.78%)270.40 (-3.46%)53.33 (-64.96%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)605.10 (n/a)452.76 (n/a)490.80 (n/a)280.10 (n/a)152.20 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (-51.95%)0.03 (-40.71%)0.02 (-46.97%)0.02 (-10.34%)0.01 (-67.17%)587.60 (+11.54%)477.36 (+41.73%)502.00 (+88.58%)288.60 (+108.07%)111.82 (-32.99%)
e216c2f — 2026-07-13 21:54:420.09 (n/a)0.05 (n/a)0.05 (n/a)0.02 (n/a)0.03 (n/a)526.80 (n/a)336.82 (n/a)266.20 (n/a)138.70 (n/a)166.86 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (-36.74%)0.03 (-32.93%)0.03 (-39.28%)0.02 (+0.81%)0.01 (-45.72%)659.40 (-0.80%)464.74 (+35.92%)481.00 (+64.67%)299.10 (+58.09%)142.04 (-23.51%)
e216c2f — 2026-07-13 21:54:420.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.02 (n/a)664.70 (n/a)341.92 (n/a)292.10 (n/a)189.20 (n/a)185.69 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (+10.83%)0.02 (+7.31%)0.01 (+22.79%)0.01 (-36.69%)0.01 (+33.03%)810.20 (+57.96%)439.40 (+5.55%)401.20 (-18.55%)196.20 (-9.75%)239.06 (+85.81%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)512.90 (n/a)416.30 (n/a)492.60 (n/a)217.40 (n/a)128.66 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (-3.44%)0.01 (+1.87%)0.01 (-18.07%)0.01 (+246.14%)0.01 (-22.67%)588.40 (-71.11%)489.18 (-35.44%)544.40 (+22.06%)220.60 (+3.57%)154.01 (-79.16%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)0.01 (n/a)2036.60 (n/a)757.66 (n/a)446.00 (n/a)213.00 (n/a)738.89 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (+20.70%)0.01 (-4.66%)0.01 (-15.93%)0.00 (-75.91%)0.01 (+193.70%)2013.40 (+315.05%)728.80 (+72.22%)517.70 (+18.93%)258.00 (-17.15%)731.14 (+936.11%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)485.10 (n/a)423.18 (n/a)435.30 (n/a)311.40 (n/a)70.57 (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)
6ff8ed1 — 2026-07-14 16:24:460.01 (-32.50%)0.01 (-43.46%)0.01 (-46.65%)0.01 (-56.59%)0.00 (-3.98%)970.70 (+130.35%)575.30 (+89.36%)517.10 (+87.42%)362.10 (+48.16%)231.30 (+235.93%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)421.40 (n/a)303.82 (n/a)275.90 (n/a)244.40 (n/a)68.85 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (-11.36%)0.01 (-10.18%)0.01 (+1.29%)0.01 (-31.77%)0.00 (-0.09%)809.20 (+46.57%)464.72 (+16.83%)418.60 (-1.27%)291.10 (+12.83%)202.34 (+77.44%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)552.10 (n/a)397.76 (n/a)424.00 (n/a)258.00 (n/a)114.03 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (+11.70%)0.01 (-3.08%)0.01 (-10.37%)0.01 (+0.82%)0.00 (+44.41%)571.50 (-0.82%)494.44 (+5.08%)520.90 (+11.57%)329.20 (-10.47%)98.37 (+26.05%)
e216c2f — 2026-07-13 21:54:420.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)576.20 (n/a)470.52 (n/a)466.90 (n/a)367.70 (n/a)78.04 (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)
6ff8ed1 — 2026-07-14 16:24:460.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)601.30 (n/a)384.02 (n/a)358.60 (n/a)268.10 (n/a)132.33 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)497.10 (n/a)343.82 (n/a)305.60 (n/a)275.30 (n/a)89.64 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)1108.00 (n/a)592.22 (n/a)606.00 (n/a)294.70 (n/a)331.72 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)566.70 (n/a)353.94 (n/a)298.70 (n/a)280.80 (n/a)120.24 (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)
6ff8ed1 — 2026-07-14 16:24:460.05 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)596.80 (n/a)377.86 (n/a)294.80 (n/a)270.10 (n/a)139.64 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)540.30 (n/a)456.48 (n/a)497.40 (n/a)294.60 (n/a)97.70 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)529.90 (n/a)362.18 (n/a)344.90 (n/a)183.20 (n/a)140.82 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)0.01 (n/a)2028.70 (n/a)756.18 (n/a)502.40 (n/a)238.50 (n/a)720.64 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)531.20 (n/a)413.40 (n/a)424.70 (n/a)241.40 (n/a)118.18 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)561.70 (n/a)444.86 (n/a)520.60 (n/a)186.60 (n/a)152.50 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)598.00 (n/a)532.68 (n/a)543.00 (n/a)452.00 (n/a)61.02 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)527.20 (n/a)416.32 (n/a)441.20 (n/a)284.40 (n/a)95.83 (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)
6ff8ed1 — 2026-07-14 16:24:460.45 (+20.95%)0.42 (+26.11%)0.41 (+14.75%)0.39 (+82.37%)0.03 (-60.91%)566.60 (-45.17%)531.06 (-23.88%)539.70 (-12.87%)490.90 (-17.32%)32.30 (-82.82%)19.22 (+20.95%)17.82 (+26.11%)17.48 (+14.75%)16.66 (+82.37%)1.10 (-60.91%)
e216c2f — 2026-07-13 21:54:420.37 (n/a)0.33 (n/a)0.36 (n/a)0.21 (n/a)0.07 (n/a)1033.30 (n/a)697.62 (n/a)619.40 (n/a)593.70 (n/a)188.04 (n/a)15.89 (n/a)14.13 (n/a)15.24 (n/a)9.13 (n/a)2.81 (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)
6ff8ed1 — 2026-07-14 16:24:460.47 (-6.05%)0.41 (-4.01%)0.41 (-0.96%)0.36 (+0.84%)0.05 (-15.64%)610.20 (-0.84%)540.22 (+3.82%)540.40 (+0.97%)467.50 (+6.44%)63.89 (-9.60%)20.19 (-6.05%)17.67 (-4.01%)17.46 (-0.96%)15.47 (+0.84%)2.11 (-15.64%)
e216c2f — 2026-07-13 21:54:420.50 (n/a)0.43 (n/a)0.41 (n/a)0.36 (n/a)0.06 (n/a)615.40 (n/a)520.34 (n/a)535.20 (n/a)439.20 (n/a)70.67 (n/a)21.49 (n/a)18.41 (n/a)17.63 (n/a)15.34 (n/a)2.50 (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)
6ff8ed1 — 2026-07-14 16:24:460.31 (+0.54%)0.31 (-0.07%)0.30 (-0.71%)0.30 (-0.63%)0.01 (+68.09%)84270.80 (+0.64%)82502.90 (+0.08%)83007.40 (+0.72%)80945.10 (-0.54%)1428.79 (+67.52%)212.24 (+0.55%)208.28 (-0.07%)206.97 (-0.71%)203.87 (-0.63%)3.61 (+68.09%)
e216c2f — 2026-07-13 21:54:420.31 (n/a)0.31 (n/a)0.31 (n/a)0.30 (n/a)0.00 (n/a)83736.50 (n/a)82433.08 (n/a)82414.90 (n/a)81386.30 (n/a)852.92 (n/a)211.09 (n/a)208.43 (n/a)208.46 (n/a)205.17 (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)
6ff8ed1 — 2026-07-14 16:24:461.03 (+0.39%)1.01 (-0.41%)1.01 (-0.68%)0.99 (-1.17%)0.02 (+41.32%)25484.60 (+1.18%)24866.80 (+0.43%)24821.90 (+0.69%)24398.20 (-0.39%)392.96 (+42.67%)704.15 (+0.39%)691.01 (-0.41%)692.12 (-0.68%)674.13 (-1.17%)10.83 (+41.32%)
e216c2f — 2026-07-13 21:54:421.03 (n/a)1.02 (n/a)1.02 (n/a)1.00 (n/a)0.01 (n/a)25186.40 (n/a)24761.20 (n/a)24652.60 (n/a)24493.90 (n/a)275.44 (n/a)701.39 (n/a)693.89 (n/a)696.88 (n/a)682.11 (n/a)7.66 (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)
6ff8ed1 — 2026-07-14 16:24:463.41 (-10.63%)2.24 (-14.78%)1.77 (-16.98%)1.12 (-29.97%)1.06 (+8.34%)7213.90 (+42.79%)4348.00 (+26.71%)4566.50 (+20.45%)2365.20 (+11.90%)2035.36 (+66.02%)893.78 (-10.63%)586.13 (-14.78%)462.92 (-16.98%)293.03 (-29.97%)277.90 (+8.34%)
e216c2f — 2026-07-13 21:54:423.81 (n/a)2.62 (n/a)2.13 (n/a)1.60 (n/a)0.98 (n/a)5052.20 (n/a)3431.40 (n/a)3791.10 (n/a)2113.70 (n/a)1225.96 (n/a)1000.11 (n/a)687.77 (n/a)557.60 (n/a)418.42 (n/a)256.51 (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)
6ff8ed1 — 2026-07-14 16:24:460.28 (-5.55%)0.23 (-6.39%)0.22 (-4.61%)0.19 (+7.26%)0.03 (-23.53%)6541.30 (-6.77%)5625.26 (+5.44%)5639.30 (+4.83%)4433.20 (+5.88%)792.11 (-26.32%)15.14 (-5.55%)12.14 (-6.39%)11.90 (-4.61%)10.26 (+7.26%)1.86 (-23.53%)
e216c2f — 2026-07-13 21:54:420.30 (n/a)0.24 (n/a)0.23 (n/a)0.18 (n/a)0.05 (n/a)7016.30 (n/a)5334.86 (n/a)5379.50 (n/a)4187.00 (n/a)1075.07 (n/a)16.03 (n/a)12.97 (n/a)12.47 (n/a)9.56 (n/a)2.43 (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)
6ff8ed1 — 2026-07-14 16:24:460.14 (+1.69%)0.10 (-11.03%)0.11 (-9.67%)0.06 (+1.28%)0.03 (+13.25%)0.14 (+1.69%)0.10 (-11.03%)0.11 (-9.67%)0.06 (+1.28%)0.03 (+13.25%)
e216c2f — 2026-07-13 21:54:420.14 (n/a)0.11 (n/a)0.12 (n/a)0.06 (n/a)0.03 (n/a)0.14 (n/a)0.11 (n/a)0.12 (n/a)0.06 (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)
6ff8ed1 — 2026-07-14 16:24:463.84 (+8.10%)3.69 (+6.68%)3.70 (+6.95%)3.53 (+6.38%)0.12 (+35.39%)3.83 (+8.10%)3.68 (+6.68%)3.70 (+6.95%)3.53 (+6.38%)0.12 (+35.39%)
e216c2f — 2026-07-13 21:54:423.55 (n/a)3.46 (n/a)3.46 (n/a)3.32 (n/a)0.09 (n/a)3.55 (n/a)3.45 (n/a)3.45 (n/a)3.31 (n/a)0.09 (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)
6ff8ed1 — 2026-07-14 16:24:467.25 (-3.16%)6.11 (-3.53%)6.04 (-2.69%)4.61 (-18.91%)1.08 (+46.55%)7.24 (-3.16%)6.11 (-3.53%)6.04 (-2.69%)4.60 (-18.91%)1.07 (+46.55%)
e216c2f — 2026-07-13 21:54:427.48 (n/a)6.33 (n/a)6.21 (n/a)5.68 (n/a)0.73 (n/a)7.48 (n/a)6.33 (n/a)6.21 (n/a)5.68 (n/a)0.73 (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)
6ff8ed1 — 2026-07-14 16:24:4614.27 (+2.92%)10.84 (+1.17%)9.81 (-6.97%)8.08 (+6.79%)3.04 (+3.71%)14.26 (+2.92%)10.83 (+1.17%)9.81 (-6.97%)8.07 (+6.79%)3.04 (+3.71%)
e216c2f — 2026-07-13 21:54:4213.86 (n/a)10.71 (n/a)10.55 (n/a)7.56 (n/a)2.93 (n/a)13.85 (n/a)10.71 (n/a)10.54 (n/a)7.56 (n/a)2.93 (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)
6ff8ed1 — 2026-07-14 16:24:463.96 (+0.58%)3.70 (-0.50%)3.73 (-1.94%)3.35 (+2.42%)0.22 (-15.68%)3.95 (+0.58%)3.70 (-0.50%)3.72 (-1.94%)3.35 (+2.42%)0.22 (-15.68%)
e216c2f — 2026-07-13 21:54:423.93 (n/a)3.72 (n/a)3.80 (n/a)3.27 (n/a)0.26 (n/a)3.93 (n/a)3.72 (n/a)3.80 (n/a)3.27 (n/a)0.26 (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)
6ff8ed1 — 2026-07-14 16:24:467.44 (-0.48%)6.14 (-7.56%)6.23 (-8.04%)4.40 (-23.07%)1.26 (+87.76%)7.44 (-0.48%)6.13 (-7.56%)6.22 (-8.04%)4.40 (-23.07%)1.26 (+87.76%)
e216c2f — 2026-07-13 21:54:427.48 (n/a)6.64 (n/a)6.77 (n/a)5.72 (n/a)0.67 (n/a)7.48 (n/a)6.63 (n/a)6.77 (n/a)5.71 (n/a)0.67 (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)
6ff8ed1 — 2026-07-14 16:24:4614.04 (+25.53%)9.39 (-4.71%)8.62 (-19.03%)7.67 (-4.80%)2.64 (+86.56%)14.03 (+25.53%)9.38 (-4.71%)8.62 (-19.03%)7.67 (-4.80%)2.64 (+86.56%)
e216c2f — 2026-07-13 21:54:4211.18 (n/a)9.85 (n/a)10.65 (n/a)8.06 (n/a)1.41 (n/a)11.18 (n/a)9.84 (n/a)10.64 (n/a)8.06 (n/a)1.41 (n/a)

test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
6ff8ed1 — 2026-07-14 16:24:463.09 (-1.78%)2.33 (-3.90%)2.79 (-1.04%)1.04 (-7.60%)0.92 (+10.33%)3.09 (-1.78%)2.33 (-3.90%)2.79 (-1.04%)1.04 (-7.60%)0.91 (+10.33%)
e216c2f — 2026-07-13 21:54:423.15 (n/a)2.42 (n/a)2.82 (n/a)1.13 (n/a)0.83 (n/a)3.14 (n/a)2.42 (n/a)2.82 (n/a)1.13 (n/a)0.83 (n/a)

test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
6ff8ed1 — 2026-07-14 16:24:460.49 (+4.98%)0.31 (-4.26%)0.45 (+28.46%)0.08 (-40.78%)0.22 (+52.62%)0.48 (+4.98%)0.31 (-4.26%)0.44 (+28.46%)0.07 (-40.78%)0.21 (+52.62%)
e216c2f — 2026-07-13 21:54:420.47 (n/a)0.33 (n/a)0.35 (n/a)0.13 (n/a)0.14 (n/a)0.46 (n/a)0.32 (n/a)0.34 (n/a)0.12 (n/a)0.14 (n/a)

test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
6ff8ed1 — 2026-07-14 16:24:460.69 (+9.73%)0.66 (+32.74%)0.68 (+16.62%)0.58 (+664.21%)0.05 (-79.81%)0.68 (+9.73%)0.65 (+32.74%)0.67 (+16.62%)0.57 (+664.21%)0.05 (-79.81%)
e216c2f — 2026-07-13 21:54:420.63 (n/a)0.50 (n/a)0.58 (n/a)0.08 (n/a)0.24 (n/a)0.62 (n/a)0.49 (n/a)0.58 (n/a)0.07 (n/a)0.24 (n/a)

test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Throughput (max)Throughput (mean)Throughput (median)Throughput (min)Throughput (stddev)
6ff8ed1 — 2026-07-14 16:24:462.61 (+15.26%)1.54 (-10.46%)1.33 (-16.62%)0.44 (-60.54%)0.98 (+118.47%)2.57 (+15.26%)1.51 (-10.46%)1.31 (-16.62%)0.43 (-60.54%)0.96 (+118.47%)
e216c2f — 2026-07-13 21:54:422.27 (n/a)1.72 (n/a)1.60 (n/a)1.11 (n/a)0.45 (n/a)2.23 (n/a)1.69 (n/a)1.57 (n/a)1.10 (n/a)0.44 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)588.00 (n/a)379.24 (n/a)299.90 (n/a)192.40 (n/a)171.22 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)568.80 (n/a)302.86 (n/a)244.30 (n/a)185.90 (n/a)153.84 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)597.70 (n/a)324.16 (n/a)277.50 (n/a)225.30 (n/a)155.49 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)494.50 (n/a)369.20 (n/a)374.80 (n/a)248.60 (n/a)115.84 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)565.90 (n/a)412.00 (n/a)417.50 (n/a)265.90 (n/a)142.85 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)535.90 (n/a)410.94 (n/a)429.70 (n/a)277.80 (n/a)106.91 (n/a)
iron/operators/leaky_relu

test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.01]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.03 (-8.59%)0.02 (-7.37%)0.03 (-12.88%)0.02 (+26.98%)0.01 (-33.88%)454.00 (-21.24%)341.76 (+1.67%)308.80 (+14.75%)269.50 (+9.37%)76.26 (-44.59%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)576.40 (n/a)336.14 (n/a)269.10 (n/a)246.40 (n/a)137.63 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.03 (-1.36%)0.03 (+7.56%)0.03 (+19.43%)0.02 (-11.13%)0.01 (+25.38%)488.40 (+12.53%)339.56 (-4.93%)313.20 (-16.28%)266.00 (+1.37%)92.10 (+44.65%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)434.00 (n/a)357.16 (n/a)374.10 (n/a)262.40 (n/a)63.67 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.04 (+8.32%)0.03 (+26.61%)0.03 (+82.54%)0.01 (-9.62%)0.01 (-2.43%)603.10 (+10.66%)323.92 (-20.73%)263.20 (-45.21%)222.40 (-7.72%)158.25 (+7.42%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)545.00 (n/a)408.62 (n/a)480.40 (n/a)241.00 (n/a)147.32 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.04 (+4.12%)0.03 (+37.37%)0.03 (+80.21%)0.00 (-0.73%)0.01 (+12.46%)2011.80 (+0.73%)607.48 (-15.03%)267.40 (-44.51%)233.90 (-3.98%)785.28 (+7.95%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)0.01 (n/a)1997.20 (n/a)714.94 (n/a)481.90 (n/a)243.60 (n/a)727.43 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.05 (+89.33%)0.03 (+42.71%)0.03 (+53.95%)0.01 (-6.22%)0.01 (+230.52%)560.30 (+6.64%)343.24 (-17.62%)263.90 (-35.03%)161.00 (-47.18%)165.10 (+94.21%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)525.40 (n/a)416.66 (n/a)406.20 (n/a)304.80 (n/a)85.01 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.03 (-16.85%)0.02 (+32.27%)0.02 (+50.40%)0.02 (+210.91%)0.00 (-64.87%)421.20 (-67.84%)356.76 (-43.02%)358.10 (-33.51%)275.70 (+20.29%)52.74 (-86.97%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)1309.50 (n/a)626.08 (n/a)538.60 (n/a)229.20 (n/a)404.72 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.04 (+76.21%)0.03 (+53.56%)0.03 (+67.81%)0.02 (+24.66%)0.01 (+203.91%)514.30 (-19.78%)358.86 (-29.92%)293.40 (-40.40%)230.30 (-43.25%)126.28 (+44.23%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)641.10 (n/a)512.08 (n/a)492.30 (n/a)405.80 (n/a)87.55 (n/a)

test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:460.03 (+24.92%)0.02 (+9.72%)0.02 (-3.15%)0.01 (-16.60%)0.01 (+82.94%)642.30 (+19.90%)412.24 (-2.69%)402.00 (+3.26%)266.10 (-19.95%)150.73 (+68.84%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)535.70 (n/a)423.62 (n/a)389.30 (n/a)332.40 (n/a)89.28 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (+3.14%)0.02 (-18.55%)0.01 (-34.94%)0.01 (-15.53%)0.01 (+24.83%)597.10 (+18.38%)495.30 (+29.05%)553.30 (+53.74%)231.60 (-3.02%)152.19 (+32.10%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)504.40 (n/a)383.80 (n/a)359.90 (n/a)238.80 (n/a)115.21 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (-5.37%)0.02 (-10.46%)0.02 (-21.29%)0.01 (-13.80%)0.01 (+8.05%)568.80 (+16.01%)413.08 (+15.68%)476.60 (+27.06%)230.70 (+5.68%)144.02 (+33.10%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)490.30 (n/a)357.08 (n/a)375.10 (n/a)218.30 (n/a)108.21 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (-1.08%)0.03 (+14.92%)0.03 (+43.45%)0.02 (-7.09%)0.01 (-2.48%)516.50 (+7.63%)329.14 (-12.46%)291.50 (-30.30%)259.60 (+1.09%)105.59 (+13.84%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)479.90 (n/a)376.00 (n/a)418.20 (n/a)256.80 (n/a)92.75 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (-28.00%)0.02 (-39.24%)0.02 (-42.88%)0.01 (-52.92%)0.01 (-29.90%)1093.30 (+112.41%)582.78 (+72.52%)505.10 (+75.08%)277.90 (+38.88%)308.82 (+110.81%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)514.70 (n/a)337.80 (n/a)288.50 (n/a)200.10 (n/a)146.49 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (-5.48%)0.02 (-5.53%)0.02 (-4.41%)0.02 (+6.73%)0.01 (-15.16%)532.80 (-6.30%)411.96 (+2.88%)463.70 (+4.63%)247.30 (+5.82%)124.86 (-12.57%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)568.60 (n/a)400.42 (n/a)443.20 (n/a)233.70 (n/a)142.81 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (+3.78%)0.02 (+10.12%)0.02 (+15.89%)0.02 (+19.64%)0.00 (-12.57%)542.00 (-16.41%)465.26 (-10.42%)465.70 (-13.71%)346.00 (-3.65%)74.61 (-28.30%)
e216c2f — 2026-07-13 21:54:420.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)648.40 (n/a)519.36 (n/a)539.70 (n/a)359.10 (n/a)104.07 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (-12.26%)0.02 (-28.90%)0.02 (-42.69%)0.01 (-40.34%)0.01 (+75.76%)619.80 (+67.60%)437.28 (+55.75%)454.50 (+74.47%)268.60 (+14.01%)164.45 (+210.90%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.00 (n/a)369.80 (n/a)280.76 (n/a)260.50 (n/a)235.60 (n/a)52.90 (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)
6ff8ed1 — 2026-07-14 16:24:460.05 (+2.43%)0.03 (+14.74%)0.03 (+19.42%)0.02 (+7.94%)0.01 (+0.18%)553.70 (-7.35%)399.72 (-14.25%)453.30 (-16.26%)245.70 (-2.34%)132.10 (-15.43%)
e216c2f — 2026-07-13 21:54:420.05 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)597.60 (n/a)466.14 (n/a)541.30 (n/a)251.60 (n/a)156.20 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (-3.55%)0.02 (-1.90%)0.02 (-4.36%)0.01 (-5.21%)0.01 (+4.17%)578.30 (+5.51%)434.02 (+3.62%)516.10 (+4.56%)256.90 (+3.67%)160.38 (+10.14%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)548.10 (n/a)418.86 (n/a)493.60 (n/a)247.80 (n/a)145.61 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (-35.94%)0.03 (-19.89%)0.03 (-10.94%)0.02 (-2.90%)0.01 (-46.81%)563.40 (+3.00%)392.72 (+13.98%)335.90 (+12.27%)231.30 (+56.07%)139.15 (-8.89%)
e216c2f — 2026-07-13 21:54:420.07 (n/a)0.04 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)547.00 (n/a)344.54 (n/a)299.20 (n/a)148.20 (n/a)152.73 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (-5.07%)0.03 (+3.29%)0.03 (+0.55%)0.02 (+36.24%)0.01 (-9.62%)503.90 (-26.61%)348.22 (-8.13%)283.00 (-0.56%)247.60 (+5.32%)123.42 (-33.04%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)686.60 (n/a)379.02 (n/a)284.60 (n/a)235.10 (n/a)184.33 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (-2.89%)0.03 (-6.12%)0.02 (-30.38%)0.02 (+20.58%)0.01 (-4.82%)487.00 (-17.06%)385.98 (+4.35%)464.50 (+43.63%)241.20 (+2.99%)121.57 (-16.96%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)587.20 (n/a)369.90 (n/a)323.40 (n/a)234.20 (n/a)146.40 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (+7.23%)0.02 (+4.04%)0.03 (+13.00%)0.01 (-7.14%)0.01 (+56.95%)612.30 (+7.69%)382.02 (+4.55%)290.00 (-11.50%)242.00 (-6.71%)173.55 (+45.44%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)568.60 (n/a)365.40 (n/a)327.70 (n/a)259.40 (n/a)119.32 (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)
6ff8ed1 — 2026-07-14 16:24:460.02 (-19.41%)0.02 (-10.89%)0.02 (-9.80%)0.01 (+1.00%)0.00 (-32.46%)638.20 (-0.99%)528.56 (+10.48%)515.40 (+10.86%)444.00 (+24.09%)83.90 (-19.39%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)644.60 (n/a)478.40 (n/a)464.90 (n/a)357.80 (n/a)104.07 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (+6.88%)0.02 (-6.96%)0.02 (-9.17%)0.01 (-14.12%)0.01 (+28.61%)553.40 (+16.43%)422.20 (+12.28%)467.60 (+10.10%)242.50 (-6.44%)134.31 (+43.61%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)475.30 (n/a)376.04 (n/a)424.70 (n/a)259.20 (n/a)93.53 (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)
6ff8ed1 — 2026-07-14 16:24:460.04 (-3.02%)0.02 (-27.42%)0.02 (-44.74%)0.01 (+8.25%)0.01 (-6.48%)622.10 (-7.63%)507.84 (+34.36%)573.10 (+80.96%)231.70 (+3.12%)158.06 (-14.29%)
e216c2f — 2026-07-13 21:54:420.04 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)673.50 (n/a)377.96 (n/a)316.70 (n/a)224.70 (n/a)184.43 (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)
6ff8ed1 — 2026-07-14 16:24:460.03 (+10.47%)0.02 (+34.63%)0.02 (+51.60%)0.01 (+77.36%)0.01 (+6.34%)560.60 (-43.62%)408.94 (-29.32%)359.90 (-34.04%)293.20 (-9.48%)130.64 (-47.68%)
e216c2f — 2026-07-13 21:54:420.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)994.30 (n/a)578.58 (n/a)545.60 (n/a)323.90 (n/a)249.69 (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)
6ff8ed1 — 2026-07-14 16:24:460.35 (+17.38%)0.24 (+0.43%)0.19 (-18.81%)0.16 (-10.50%)0.09 (+86.84%)606.20 (+11.74%)449.36 (+6.62%)506.60 (+23.17%)280.40 (-14.80%)148.19 (+73.21%)
e216c2f — 2026-07-13 21:54:420.30 (n/a)0.24 (n/a)0.24 (n/a)0.18 (n/a)0.05 (n/a)542.50 (n/a)421.44 (n/a)411.30 (n/a)329.10 (n/a)85.56 (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)
6ff8ed1 — 2026-07-14 16:24:460.36 (+1.25%)0.26 (-0.75%)0.33 (+7.07%)0.09 (-42.49%)0.11 (+19.94%)1060.90 (+73.89%)481.08 (+15.64%)298.50 (-6.60%)274.80 (-1.22%)334.64 (+101.17%)
e216c2f — 2026-07-13 21:54:420.35 (n/a)0.27 (n/a)0.31 (n/a)0.16 (n/a)0.09 (n/a)610.10 (n/a)416.02 (n/a)319.60 (n/a)278.20 (n/a)166.35 (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)
6ff8ed1 — 2026-07-14 16:24:460.35 (-28.36%)0.22 (-18.09%)0.19 (-10.41%)0.17 (+8.23%)0.08 (-43.17%)586.50 (-7.61%)483.56 (+12.12%)530.90 (+11.60%)284.00 (+39.56%)127.75 (-24.10%)
e216c2f — 2026-07-13 21:54:420.48 (n/a)0.27 (n/a)0.21 (n/a)0.15 (n/a)0.13 (n/a)634.80 (n/a)431.30 (n/a)475.70 (n/a)203.50 (n/a)168.32 (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)
6ff8ed1 — 2026-07-14 16:24:460.29 (+25.64%)0.18 (+6.60%)0.18 (+17.07%)0.04 (-67.54%)0.10 (+108.66%)1948.40 (+208.10%)693.76 (+45.85%)420.10 (-14.58%)253.10 (-20.38%)709.24 (+470.54%)
e216c2f — 2026-07-13 21:54:420.23 (n/a)0.16 (n/a)0.15 (n/a)0.12 (n/a)0.05 (n/a)632.40 (n/a)475.66 (n/a)491.80 (n/a)317.90 (n/a)124.31 (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)
6ff8ed1 — 2026-07-14 16:24:460.30 (+57.10%)0.22 (+38.36%)0.20 (+27.85%)0.17 (+32.46%)0.05 (+75.96%)441.10 (-24.51%)348.64 (-26.71%)378.00 (-21.79%)246.40 (-36.35%)76.16 (-13.84%)
e216c2f — 2026-07-13 21:54:420.19 (n/a)0.16 (n/a)0.15 (n/a)0.13 (n/a)0.03 (n/a)584.30 (n/a)475.72 (n/a)483.30 (n/a)387.10 (n/a)88.40 (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)
6ff8ed1 — 2026-07-14 16:24:460.31 (+15.30%)0.17 (-11.04%)0.14 (-3.56%)0.11 (-13.07%)0.08 (+15.55%)651.40 (+15.03%)500.70 (+15.53%)537.70 (+3.68%)235.00 (-13.28%)164.54 (+13.19%)
e216c2f — 2026-07-13 21:54:420.27 (n/a)0.19 (n/a)0.14 (n/a)0.13 (n/a)0.07 (n/a)566.30 (n/a)433.40 (n/a)518.60 (n/a)271.00 (n/a)145.36 (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)
6ff8ed1 — 2026-07-14 16:24:460.54 (+21.51%)0.29 (+6.81%)0.24 (-14.52%)0.20 (+275.42%)0.14 (-0.12%)648.00 (-73.37%)514.26 (-37.52%)550.00 (+17.00%)244.00 (-17.71%)158.55 (-82.45%)
e216c2f — 2026-07-13 21:54:420.44 (n/a)0.27 (n/a)0.28 (n/a)0.05 (n/a)0.14 (n/a)2432.90 (n/a)823.14 (n/a)470.10 (n/a)296.50 (n/a)903.28 (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)
6ff8ed1 — 2026-07-14 16:24:460.45 (-12.48%)0.26 (-27.33%)0.21 (-38.34%)0.20 (+3.86%)0.11 (-20.07%)666.60 (-3.71%)555.02 (+32.69%)612.10 (+62.19%)292.90 (+14.28%)151.04 (-14.67%)
e216c2f — 2026-07-13 21:54:420.51 (n/a)0.36 (n/a)0.35 (n/a)0.19 (n/a)0.13 (n/a)692.30 (n/a)418.28 (n/a)377.40 (n/a)256.30 (n/a)177.01 (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)
6ff8ed1 — 2026-07-14 16:24:460.39 (-10.39%)0.26 (-17.86%)0.23 (-10.79%)0.21 (-1.63%)0.07 (-25.83%)629.80 (+1.66%)540.42 (+18.24%)574.40 (+12.10%)338.40 (+11.57%)115.26 (-16.13%)
e216c2f — 2026-07-13 21:54:420.43 (n/a)0.31 (n/a)0.26 (n/a)0.21 (n/a)0.10 (n/a)619.50 (n/a)457.04 (n/a)512.40 (n/a)303.30 (n/a)137.43 (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)
6ff8ed1 — 2026-07-14 16:24:460.00 (-50.00%)0.00 (-14.29%)0.00 (+0.00%)0.00 (+0.00%)0.00 (-69.38%)19997.84 (-12.67%)16940.51 (-8.22%)16402.26 (-26.38%)13686.41 (+103.45%)2564.82 (-62.62%)
e216c2f — 2026-07-13 21:54:420.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)22899.76 (n/a)18458.31 (n/a)22279.89 (n/a)6727.16 (n/a)6860.73 (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)
6ff8ed1 — 2026-07-14 16:24:460.00 (-14.29%)0.00 (-41.51%)0.00 (-58.33%)0.00 (+0.00%)0.00 (-13.01%)21430.84 (-6.08%)15616.74 (+55.64%)15084.09 (+119.14%)7116.83 (+17.52%)5771.91 (-19.50%)
e216c2f — 2026-07-13 21:54:420.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)22818.84 (n/a)10033.62 (n/a)6883.34 (n/a)6056.00 (n/a)7169.71 (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)
6ff8ed1 — 2026-07-14 16:24:460.13 (-12.28%)0.09 (-15.62%)0.08 (-19.88%)0.07 (-8.94%)0.03 (-9.07%)30256.62 (+9.73%)24431.98 (+19.11%)26690.02 (+24.90%)15695.61 (+14.01%)6435.86 (+17.94%)
e216c2f — 2026-07-13 21:54:420.15 (n/a)0.11 (n/a)0.10 (n/a)0.08 (n/a)0.03 (n/a)27574.44 (n/a)20512.79 (n/a)21368.30 (n/a)13767.47 (n/a)5456.68 (n/a)
iron/operators/transpose

test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:461.79 (+65.41%)1.31 (+54.00%)1.49 (+89.25%)0.72 (+16.15%)0.49 (+165.74%)724.90 (-13.91%)458.84 (-28.47%)352.30 (-47.16%)292.50 (-39.54%)196.46 (+39.95%)
e216c2f — 2026-07-13 21:54:421.08 (n/a)0.85 (n/a)0.79 (n/a)0.62 (n/a)0.18 (n/a)842.00 (n/a)641.50 (n/a)666.70 (n/a)483.80 (n/a)140.38 (n/a)

test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:463.12 (+27.00%)2.27 (+28.50%)2.30 (+2.73%)1.37 (+334.76%)0.68 (-27.02%)766.80 (-77.00%)501.64 (-53.78%)456.70 (-2.64%)335.70 (-21.25%)168.55 (-86.68%)
e216c2f — 2026-07-13 21:54:422.46 (n/a)1.76 (n/a)2.24 (n/a)0.31 (n/a)0.93 (n/a)3333.70 (n/a)1085.28 (n/a)469.10 (n/a)426.30 (n/a)1264.95 (n/a)

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)
4bb8427 — 2026-06-25 20:01:371.58 (-2.67%)1.21 (+5.65%)1.29 (+41.89%)0.66 (-20.05%)0.38 (-2.53%)792.90 (+25.08%)479.12 (-3.80%)406.80 (-29.52%)331.10 (+2.76%)189.18 (+26.62%)
4bb8427 — 2026-06-23 22:46:491.63 (n/a)1.14 (n/a)0.91 (n/a)0.83 (n/a)0.38 (n/a)633.90 (n/a)498.02 (n/a)577.20 (n/a)322.20 (n/a)149.40 (n/a)

test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
6ff8ed1 — 2026-07-14 16:24:461.76 (+16.63%)1.07 (-0.71%)0.90 (-4.93%)0.88 (+39.04%)0.38 (+6.58%)599.00 (-28.07%)524.92 (-1.81%)582.30 (+5.18%)297.90 (-14.27%)128.34 (-33.65%)
e216c2f — 2026-07-13 21:54:421.51 (n/a)1.08 (n/a)0.95 (n/a)0.63 (n/a)0.36 (n/a)832.80 (n/a)534.60 (n/a)553.60 (n/a)347.50 (n/a)193.43 (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)
4bb8427 — 2026-06-25 20:01:371.75 (-15.44%)1.32 (-0.50%)1.36 (+22.47%)0.77 (-19.20%)0.36 (-21.75%)678.10 (+23.76%)430.02 (-0.09%)385.10 (-18.34%)300.10 (+18.29%)146.96 (+19.46%)
4bb8427 — 2026-06-23 22:46:492.07 (n/a)1.32 (n/a)1.11 (n/a)0.96 (n/a)0.46 (n/a)547.90 (n/a)430.40 (n/a)471.60 (n/a)253.70 (n/a)123.02 (n/a)
Phoenix - Examples

IRON

Tested on 2026_07_14_16_31_05 at commit 6ff8ed1.

Trends:

IRON Trends

Two opt-in build-ergonomics knobs on the aiecc compilation rules; both
default to current behavior so nothing changes unless the env var is set.

- AIECC_PATH: point the compilation rules at a locally-built aiecc (e.g. a
  compiler under development) instead of the one under mlir_aie_dir/bin,
  without having to replace the installed binary.
- AIECC_JOBS: pass -j<N> to aiecc instead of the hardcoded -j1, so multi-core
  hosts can compile the AIE cores in parallel. Default '1' = unchanged.
@atassis atassis force-pushed the feat/aiecc-path-jobs-env branch from 6563352 to 1e0d6c9 Compare July 13, 2026 22:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants