Skip to content

compilation: stop pinning aiecc to one thread - #185

Open
atassis wants to merge 1 commit into
amd:develfrom
atassis:pr/aiecc-jobs-default
Open

compilation: stop pinning aiecc to one thread#185
atassis wants to merge 1 commit into
amd:develfrom
atassis:pr/aiecc-jobs-default

Conversation

@atassis

@atassis atassis commented Sep 6, 2026

Copy link
Copy Markdown
Contributor

Problem

Both aiecc rules default AIECC_JOBS to 1:

options = [f"-j{os.environ.get('AIECC_JOBS', '1')}"]

so every design's per-core compiles run one at a time, whatever the machine has. aiecc's own default
is auto-detect, and nothing here explains the override.

Fix

Default to 0, which is aiecc's auto-detect. The environment variable still overrides, so anyone who
wants one thread keeps it.

Test

On a 24-core design (StaticMHA, 20 heads, 8 pipelines): aiecc goes from 7.7 s to 6.0 s. Nothing
above 8 threads helps.

-j does not change what aiecc produces, so this is safe rather than a trade: insts.bin and all 24
per-core ELFs are byte-identical between -j1 and -j16, and input_with_addresses.mlir differs
only in the work-dir path it embeds — which two runs at the same -j also differ in.

Both aiecc rules defaulted AIECC_JOBS to 1, so every design's per-core compiles
ran serially. On the encoder-MHA design (24 cores) that is 7.7 s against 6.0 s
at aiecc's own auto-detect; nothing above 8 helps. -j does not change the
output: insts.bin and all 24 per-core ELFs are byte-identical between -j1 and
-j16.

@andrej andrej left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is anything setting AIECC_JOBS = 1 for CI? If not, please add that, our our CI will constantly OOM.

Comment on lines +559 to +564
# aiecc's own default. "1" here made every design's per-core compiles serial: on
# the encoder-MHA design (24 cores) aiecc costs 7.7 s at -j1 and 6.0 s at -j0, and
# nothing above 8 helps. Safe because -j does not change what aiecc produces --
# measured on that design, insts.bin and all 24 per-core ELFs are byte-identical
# between -j1 and -j16, and input_with_addresses.mlir differs only in the work-dir
# path it embeds, which two runs at the SAME -j also differ in.

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
# aiecc's own default. "1" here made every design's per-core compiles serial: on
# the encoder-MHA design (24 cores) aiecc costs 7.7 s at -j1 and 6.0 s at -j0, and
# nothing above 8 helps. Safe because -j does not change what aiecc produces --
# measured on that design, insts.bin and all 24 per-core ELFs are byte-identical
# between -j1 and -j16, and input_with_addresses.mlir differs only in the work-dir
# path it embeds, which two runs at the SAME -j also differ in.
# Default parallelism setting used when environment variable AIECC_JOBS is not set.
# 0 means auto-detect.

@github-actions

github-actions Bot commented Sep 8, 2026

Copy link
Copy Markdown
Contributor

CI Test Results

b13d03d (2026_09_08_22_57_11)

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] - - 480.46
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0] - - 429.14
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0] - - 470.50
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] - - 306.96
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32] - - 282.32
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32] - - 376.90
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32] - - 466.04
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32] - - 548.72
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32] - - 498.32
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] - - 454.70
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024] - - 417.98
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512] - - 408.94
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] - - 412.26
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024] - - 666.90
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512] - - 424.58
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] - - 368.16
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 350.92
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 497.30
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 446.88
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 824.96
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 448.88
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] - - 515.28
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] - - 571.98
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] - - 82973.24
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] - - 24666.74
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] - - 4699.64
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] - - 6707.78
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] - - 0.09
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048] - - 3.51
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024] - - 6.61
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512] - - 9.01
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024] - - 3.72
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024] - - 6.71
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024] - - 9.49
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2] - - 2.52
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2] - - 0.40
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4] - - 0.62
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8] - - 1.57
test_gemv_gelu[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128] - - -
test_gemv_gelu[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048] - - -
test_gemv_gelu[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024] - - -
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] - - 290.46
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 430.96
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 437.90
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 453.86
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 380.08
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 466.52
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] - - 328.26
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1] - - 314.92
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25] - - 357.24
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01] - - 424.28
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01] - - 357.56
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01] - - 431.80
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01] - - 416.02
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01] - - 468.82
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] - - 468.86
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024] - - 423.48
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024] - - 456.66
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512] - - 443.92
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512] - - 552.90
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256] - - 504.84
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] - - 423.14
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 474.96
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 433.98
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 427.36
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 482.80
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 411.48
iron/operators/repeat
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_cols_without_a_legal_split_is_rejected[cols_1031-why_prime > 1023: the only divisors are 1 and cols, neither legal] - - -
test_cols_without_a_legal_split_is_rejected[cols_2062-why_2 x 1031: the only word-aligned chunk leaves a 1031-wide chunk count] - - -
test_cols_without_a_legal_split_is_rejected[cols_513-why_odd: every divisor is odd, so no chunk is a whole 32-bit word] - - -
test_repeat[rows_4-cols_1024-repeat_2-transfer_size_None] - - 415.12
test_repeat[rows_8-cols_512-repeat_4-transfer_size_64] - - 363.54
test_repeat[rows_8-cols_64-repeat_4-transfer_size_None] - - 377.92
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] - - 471.94
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True] - - 370.44
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False] - - 328.50
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True] - - 405.52
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False] - - 520.86
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True] - - 498.98
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False] - - 668.42
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True] - - 703.92
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False] - - 514.62
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True] - - 474.80
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False] - - 422.04
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] - - 379.46
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0] - - 460.34
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0] - - 462.12
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0] - - 434.78
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0] - - 395.12
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0] - - 686.36
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] - - 381.60
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 409.46
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 352.12
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 863.84
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 414.74
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 729.48
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] - - 442.08
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 685.46
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 339.50
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] - - 434.58
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048] - - 421.58
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512] - - 418.56
iron/operators/strided_copy
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_strided_copy[chunked_transfer] - - 313.44
test_strided_copy[contiguous] - - 323.56
test_strided_copy[four_channels] - - 434.52
test_strided_copy[kv_slot0] - - 641.02
test_strided_copy[kv_slot5] - - 354.10
test_strided_copy[kv_slot5_four_channels] - - 501.52
test_strided_copy[kv_slot5_two_channels] - - 438.12
test_strided_copy[kv_slot_last] - - 470.34
test_strided_copy[two_channels] - - 395.00
test_strided_copy[two_channels_chunked] - - 417.54
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter0] - - -
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter1] - - -
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter2] - - -
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter3] - - -
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter4] - - -
iron/operators/swiglu_decode
Test Krackan Status Krackan Phoenix Status Phoenix Latency (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584] - - 16907.58
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048] - - 15235.10
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] - - 23580.53
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] - - 343.80
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] - - 389.36
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] - - 410.50
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] - - 491.46
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] - - 453.48
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] - - 802.94
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] - - 527.28
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2] - - 1133.26
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1] - - 429.28
Krackan - Examples

IRON

Tested on 2026_09_08_22_57_11 at commit b13d03d.

iron/applications/llama_3.2_1b
TestChecksTTFT (mean)TPS (mean)
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_1]✅ 5/52.06n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_40]✅ 5/52.107.82
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_1]✅ 5/52.01n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_40]✅ 5/52.027.60

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)
b13d03d — 2026-09-08 22:52:072.08 (-0.05%)2.06 (-0.48%)2.05 (-0.92%)2.03 (-0.93%)0.02 (+39.16%)
deb6e1e — 2026-09-03 22:57:172.09 (n/a)2.07 (n/a)2.07 (n/a)2.05 (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)
b13d03d — 2026-09-08 22:52:077.85 (-0.01%)7.82 (-0.05%)7.82 (-0.12%)7.79 (+0.04%)0.03 (+8.27%)2.22 (+2.35%)2.10 (+0.33%)2.07 (-0.72%)2.06 (+0.39%)0.07 (+47.98%)
deb6e1e — 2026-09-03 22:57:177.85 (n/a)7.82 (n/a)7.83 (n/a)7.79 (n/a)0.02 (n/a)2.17 (n/a)2.09 (n/a)2.09 (n/a)2.05 (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)
b13d03d — 2026-09-08 22:52:072.03 (-0.44%)2.01 (-0.65%)2.02 (-0.54%)2.00 (-0.84%)0.02 (+49.92%)
deb6e1e — 2026-09-03 22:57:172.04 (n/a)2.03 (n/a)2.03 (n/a)2.01 (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)
b13d03d — 2026-09-08 22:52:077.65 (+0.25%)7.60 (-0.10%)7.59 (-0.18%)7.56 (-0.28%)0.03 (+87.57%)2.04 (-0.58%)2.02 (-0.30%)2.02 (-0.64%)2.01 (+0.35%)0.02 (-15.67%)
deb6e1e — 2026-09-03 22:57:177.63 (n/a)7.61 (n/a)7.61 (n/a)7.58 (n/a)0.02 (n/a)2.05 (n/a)2.03 (n/a)2.03 (n/a)2.00 (n/a)0.02 (n/a)
Phoenix - Small

IRON

Tested on 2026_09_08_22_26_55 at commit b13d03d.

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]✅ 5/5480.460.03n/a
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]✅ 5/5429.140.03n/a
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]✅ 5/5470.500.03n/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]✅ 5/5306.960.02n/a
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]✅ 5/5282.320.02n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]✅ 5/5376.900.02n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]✅ 5/5466.040.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]✅ 5/5548.720.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]✅ 5/5498.320.01n/a
iron/operators/elementwise_add
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5454.700.03n/a
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5417.980.03n/a
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5408.940.03n/a
iron/operators/elementwise_mul
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5412.260.04n/a
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5666.900.03n/a
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5424.580.03n/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]✅ 5/5368.160.02n/a
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5350.920.03n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5497.300.02n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5446.880.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5824.960.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5448.880.02n/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]✅ 5/5515.280.4418.64
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]✅ 5/5571.980.4117.47
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]✅ 5/582973.240.30207.07
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]✅ 5/524666.741.02696.53
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]✅ 5/54699.641.78466.99
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]✅ 5/56707.780.1910.15
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/a0.090.09
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a3.513.50
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]✅ 5/5n/a6.616.61
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]✅ 5/5n/a9.019.00
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a3.723.72
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a6.716.70
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a9.499.49
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2]✅ 5/5n/a2.522.51
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2]✅ 5/5n/a0.400.40
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4]✅ 5/5n/a0.620.61
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8]✅ 5/5n/a1.571.55
test_gemv_gelu[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128]❌ 0/5n/an/an/a
test_gemv_gelu[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]❌ 0/5n/an/an/a
test_gemv_gelu[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]❌ 0/5n/an/an/a
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]✅ 5/5290.460.03n/a
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5430.960.02n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5437.900.02n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5453.860.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5380.080.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5466.520.02n/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]✅ 5/5328.260.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1]✅ 5/5314.920.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25]✅ 5/5357.240.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01]✅ 5/5424.280.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01]✅ 5/5357.560.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01]✅ 5/5431.800.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01]✅ 5/5416.020.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01]✅ 5/5468.820.02n/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]✅ 5/5468.860.02n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]✅ 5/5423.480.02n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]✅ 5/5456.660.02n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]✅ 5/5443.920.02n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]✅ 5/5552.900.02n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]✅ 5/5504.840.02n/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]✅ 5/5423.140.02n/a
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5474.960.02n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5433.980.02n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5427.360.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5482.800.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5411.480.02n/a
iron/operators/repeat
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_cols_without_a_legal_split_is_rejected[cols_1031-why_prime > 1023: the only divisors are 1 and cols, neither legal]✅ 5/5n/an/an/a
test_cols_without_a_legal_split_is_rejected[cols_2062-why_2 x 1031: the only word-aligned chunk leaves a 1031-wide chunk count]✅ 5/5n/an/an/a
test_cols_without_a_legal_split_is_rejected[cols_513-why_odd: every divisor is odd, so no chunk is a whole 32-bit word]✅ 5/5n/an/an/a
test_repeat[rows_4-cols_1024-repeat_2-transfer_size_None]✅ 5/5415.120.07n/a
test_repeat[rows_8-cols_512-repeat_4-transfer_size_64]✅ 5/5363.540.12n/a
test_repeat[rows_8-cols_64-repeat_4-transfer_size_None]✅ 5/5377.920.01n/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]✅ 5/5471.940.02n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]✅ 5/5370.440.04n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]✅ 5/5328.500.03n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]✅ 5/5405.520.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]✅ 5/5520.860.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]✅ 5/5498.980.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]✅ 5/5668.420.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]✅ 5/5703.920.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]✅ 5/5514.620.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]✅ 5/5474.800.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]✅ 5/5422.040.02n/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]✅ 5/5379.460.28n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]✅ 5/5460.340.22n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]✅ 5/5462.120.25n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]✅ 5/5434.780.18n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]✅ 5/5395.120.22n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]✅ 5/5686.360.11n/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]✅ 5/5381.600.02n/a
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5409.460.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5352.120.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5863.840.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5414.740.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5729.480.02n/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]✅ 5/5442.080.02n/a
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5685.460.02n/a
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5339.500.03n/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]✅ 5/5434.580.32n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]✅ 5/5421.580.35n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5418.560.35n/a
iron/operators/strided_copy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_strided_copy[chunked_transfer]✅ 5/5313.440.01n/a
test_strided_copy[contiguous]✅ 5/5323.560.01n/a
test_strided_copy[four_channels]✅ 5/5434.520.01n/a
test_strided_copy[kv_slot0]✅ 5/5641.020.38n/a
test_strided_copy[kv_slot5]✅ 5/5354.100.40n/a
test_strided_copy[kv_slot5_four_channels]✅ 5/5501.520.27n/a
test_strided_copy[kv_slot5_two_channels]✅ 5/5438.120.33n/a
test_strided_copy[kv_slot_last]✅ 5/5470.340.33n/a
test_strided_copy[two_channels]✅ 5/5395.000.01n/a
test_strided_copy[two_channels_chunked]✅ 5/5417.540.01n/a
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter0]✅ 1/1n/an/an/a
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter1]✅ 1/1n/an/an/a
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter2]✅ 1/1n/an/an/a
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter3]✅ 1/1n/an/an/a
test_transfer_size_not_dividing_per_channel_share_is_rejected[iter4]✅ 1/1n/an/an/a
iron/operators/swiglu_decode
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584]✅ 5/516907.580.00n/a
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]✅ 5/515235.100.00n/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]✅ 5/523580.530.09n/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]✅ 5/5343.800.03n/a
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5389.360.02n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5410.500.02n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5491.460.02n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5453.480.02n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5802.940.01n/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]✅ 5/5527.281.26n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2]✅ 5/51133.261.67n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1]✅ 5/5429.281.57n/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)
b13d03d — 2026-09-08 22:22:280.03 (-30.49%)0.03 (-22.18%)0.02 (-14.57%)0.02 (+1.82%)0.00 (-62.16%)549.50 (-1.79%)480.46 (+18.32%)493.90 (+17.04%)365.30 (+43.88%)77.08 (-45.01%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)559.50 (n/a)406.08 (n/a)422.00 (n/a)253.90 (n/a)140.18 (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)
b13d03d — 2026-09-08 22:22:280.04 (-30.16%)0.03 (-12.82%)0.03 (+26.23%)0.02 (+90.89%)0.01 (-61.64%)549.70 (-47.62%)429.14 (-16.94%)429.00 (-20.79%)287.90 (+43.23%)117.30 (-66.22%)
deb6e1e — 2026-09-03 22:38:330.06 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)0.02 (n/a)1049.40 (n/a)516.68 (n/a)541.60 (n/a)201.00 (n/a)347.21 (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)
b13d03d — 2026-09-08 22:22:280.04 (-19.65%)0.03 (-16.61%)0.03 (-14.54%)0.02 (-7.50%)0.01 (-38.85%)634.50 (+8.09%)470.50 (+13.57%)450.10 (+17.00%)316.80 (+24.43%)118.47 (-20.58%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)587.00 (n/a)414.30 (n/a)384.70 (n/a)254.60 (n/a)149.17 (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)
b13d03d — 2026-09-08 22:22:280.02 (-31.07%)0.02 (-23.77%)0.02 (-16.77%)0.01 (-21.65%)0.00 (-42.78%)407.40 (+27.63%)306.96 (+28.24%)291.70 (+20.14%)221.40 (+45.09%)67.56 (+8.58%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)319.20 (n/a)239.36 (n/a)242.80 (n/a)152.60 (n/a)62.22 (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)
b13d03d — 2026-09-08 22:22:280.02 (-11.25%)0.02 (+7.92%)0.02 (-0.92%)0.01 (+30.24%)0.00 (-45.84%)376.70 (-23.22%)282.32 (-15.40%)272.30 (+0.93%)233.60 (+12.69%)58.45 (-55.87%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)490.60 (n/a)333.72 (n/a)269.80 (n/a)207.30 (n/a)132.45 (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)
b13d03d — 2026-09-08 22:22:280.02 (+2.40%)0.02 (-18.85%)0.02 (-14.81%)0.01 (-43.05%)0.01 (+139.81%)581.00 (+75.58%)376.90 (+37.15%)306.40 (+17.39%)234.20 (-2.34%)150.43 (+316.71%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)330.90 (n/a)274.80 (n/a)261.00 (n/a)239.80 (n/a)36.10 (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)
b13d03d — 2026-09-08 22:22:280.02 (-18.94%)0.01 (+4.71%)0.01 (+14.45%)0.01 (+216.35%)0.00 (-55.83%)597.30 (-68.39%)466.04 (-41.77%)493.50 (-12.62%)300.30 (+23.33%)126.42 (-81.53%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)0.01 (n/a)1889.60 (n/a)800.34 (n/a)564.80 (n/a)243.50 (n/a)684.35 (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)
b13d03d — 2026-09-08 22:22:280.03 (+22.46%)0.01 (-8.04%)0.01 (-10.58%)0.01 (-47.67%)0.01 (+64.31%)986.50 (+91.11%)548.72 (+35.11%)540.70 (+11.81%)208.20 (-18.32%)314.07 (+148.56%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)516.20 (n/a)406.14 (n/a)483.60 (n/a)254.90 (n/a)126.36 (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)
b13d03d — 2026-09-08 22:22:280.02 (-9.88%)0.01 (-23.99%)0.01 (-35.43%)0.01 (+11.81%)0.00 (-22.47%)631.70 (-10.56%)498.32 (+23.91%)512.00 (+54.87%)272.50 (+10.95%)136.88 (-27.53%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)706.30 (n/a)402.16 (n/a)330.60 (n/a)245.60 (n/a)188.89 (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)
b13d03d — 2026-09-08 22:22:280.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)602.60 (n/a)454.70 (n/a)461.50 (n/a)290.20 (n/a)112.39 (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)
b13d03d — 2026-09-08 22:22:280.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)499.10 (n/a)417.98 (n/a)474.10 (n/a)295.30 (n/a)93.23 (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)
b13d03d — 2026-09-08 22:22:280.05 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)490.70 (n/a)408.94 (n/a)479.90 (n/a)267.60 (n/a)105.23 (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)
b13d03d — 2026-09-08 22:22:280.06 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.02 (n/a)734.50 (n/a)412.26 (n/a)276.10 (n/a)216.50 (n/a)237.88 (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)
b13d03d — 2026-09-08 22:22:280.05 (n/a)0.03 (n/a)0.04 (n/a)0.01 (n/a)0.02 (n/a)2035.20 (n/a)666.90 (n/a)274.20 (n/a)242.40 (n/a)772.68 (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)
b13d03d — 2026-09-08 22:22:280.05 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)586.70 (n/a)424.58 (n/a)496.90 (n/a)256.80 (n/a)149.76 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)520.20 (n/a)368.16 (n/a)296.20 (n/a)247.00 (n/a)127.03 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)525.20 (n/a)350.92 (n/a)297.60 (n/a)238.30 (n/a)114.21 (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)
b13d03d — 2026-09-08 22:22:280.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)605.30 (n/a)497.30 (n/a)505.90 (n/a)347.00 (n/a)93.66 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)649.80 (n/a)446.88 (n/a)495.30 (n/a)248.50 (n/a)184.47 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)0.01 (n/a)2448.10 (n/a)824.96 (n/a)554.10 (n/a)274.80 (n/a)918.36 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)550.70 (n/a)448.88 (n/a)504.20 (n/a)309.00 (n/a)119.11 (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)
b13d03d — 2026-09-08 22:22:280.51 (-22.44%)0.44 (-5.82%)0.42 (-0.87%)0.35 (-4.94%)0.06 (-44.69%)630.50 (+5.21%)515.28 (+3.87%)521.80 (+0.89%)431.70 (+28.94%)77.33 (-21.54%)21.86 (-22.44%)18.64 (-5.82%)18.09 (-0.87%)14.97 (-4.94%)2.69 (-44.69%)
deb6e1e — 2026-09-03 22:38:330.66 (n/a)0.46 (n/a)0.43 (n/a)0.37 (n/a)0.11 (n/a)599.30 (n/a)496.08 (n/a)517.20 (n/a)334.80 (n/a)98.56 (n/a)28.19 (n/a)19.79 (n/a)18.25 (n/a)15.75 (n/a)4.86 (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)
b13d03d — 2026-09-08 22:22:280.52 (-5.96%)0.41 (-2.45%)0.38 (-10.37%)0.27 (-19.07%)0.11 (+17.79%)810.00 (+23.57%)571.98 (+4.92%)585.00 (+11.58%)422.60 (+6.34%)157.71 (+45.11%)22.33 (-5.96%)17.47 (-2.45%)16.13 (-10.37%)11.65 (-19.07%)4.51 (+17.79%)
deb6e1e — 2026-09-03 22:38:330.56 (n/a)0.42 (n/a)0.42 (n/a)0.34 (n/a)0.09 (n/a)655.50 (n/a)545.14 (n/a)524.30 (n/a)397.40 (n/a)108.69 (n/a)23.74 (n/a)17.91 (n/a)18.00 (n/a)14.40 (n/a)3.83 (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)
b13d03d — 2026-09-08 22:22:280.31 (-0.23%)0.30 (-0.00%)0.30 (-0.56%)0.30 (+0.90%)0.00 (-41.77%)83692.20 (-0.89%)82973.24 (-0.01%)83206.30 (+0.56%)81867.30 (+0.23%)732.88 (-42.20%)209.85 (-0.23%)207.07 (-0.00%)206.47 (-0.56%)205.27 (+0.90%)1.84 (-41.77%)
deb6e1e — 2026-09-03 22:38:330.31 (n/a)0.30 (n/a)0.30 (n/a)0.30 (n/a)0.00 (n/a)84441.40 (n/a)82982.52 (n/a)82742.70 (n/a)81682.30 (n/a)1268.00 (n/a)210.33 (n/a)207.07 (n/a)207.63 (n/a)203.45 (n/a)3.16 (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)
b13d03d — 2026-09-08 22:22:281.03 (+0.50%)1.02 (+0.73%)1.02 (+0.54%)1.01 (+0.90%)0.01 (-7.48%)25024.90 (-0.90%)24666.74 (-0.72%)24667.20 (-0.53%)24422.20 (-0.50%)232.32 (-8.89%)703.45 (+0.50%)696.53 (+0.73%)696.47 (+0.54%)686.51 (+0.90%)6.53 (-7.48%)
deb6e1e — 2026-09-03 22:38:331.03 (n/a)1.01 (n/a)1.01 (n/a)1.00 (n/a)0.01 (n/a)25251.20 (n/a)24846.02 (n/a)24799.50 (n/a)24544.40 (n/a)255.00 (n/a)699.95 (n/a)691.51 (n/a)692.75 (n/a)680.36 (n/a)7.06 (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)
b13d03d — 2026-09-08 22:22:282.29 (-37.44%)1.78 (-37.03%)1.77 (-39.43%)1.36 (-30.25%)0.39 (-54.19%)5912.90 (+43.38%)4699.64 (+52.71%)4550.30 (+65.09%)3522.00 (+59.85%)1008.33 (+4.82%)600.21 (-37.44%)466.99 (-37.03%)464.57 (-39.43%)357.51 (-30.25%)101.02 (-54.19%)
deb6e1e — 2026-09-03 22:38:333.66 (n/a)2.83 (n/a)2.92 (n/a)1.95 (n/a)0.84 (n/a)4124.00 (n/a)3077.52 (n/a)2756.20 (n/a)2203.30 (n/a)962.01 (n/a)959.42 (n/a)741.61 (n/a)766.98 (n/a)512.59 (n/a)220.50 (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)
b13d03d — 2026-09-08 22:22:280.22 (-9.15%)0.19 (-8.79%)0.18 (-12.19%)0.15 (-14.34%)0.02 (-5.28%)8170.90 (+16.74%)6707.78 (+9.84%)6748.60 (+13.89%)5653.10 (+10.07%)935.30 (+21.45%)11.87 (-9.15%)10.15 (-8.79%)9.94 (-12.19%)8.21 (-14.34%)1.34 (-5.28%)
deb6e1e — 2026-09-03 22:38:330.24 (n/a)0.21 (n/a)0.21 (n/a)0.18 (n/a)0.03 (n/a)6999.10 (n/a)6106.98 (n/a)5925.70 (n/a)5135.70 (n/a)770.08 (n/a)13.07 (n/a)11.13 (n/a)11.33 (n/a)9.59 (n/a)1.42 (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)
b13d03d — 2026-09-08 22:22:280.14 (n/a)0.09 (n/a)0.08 (n/a)0.06 (n/a)0.04 (n/a)0.13 (n/a)0.09 (n/a)0.08 (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)
b13d03d — 2026-09-08 22:22:283.90 (n/a)3.51 (n/a)3.50 (n/a)3.27 (n/a)0.26 (n/a)3.89 (n/a)3.50 (n/a)3.50 (n/a)3.26 (n/a)0.26 (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)
b13d03d — 2026-09-08 22:22:287.36 (-0.67%)6.61 (+2.60%)7.03 (+5.05%)5.67 (+0.10%)0.75 (-0.86%)7.35 (-0.67%)6.61 (+2.60%)7.03 (+5.05%)5.66 (+0.10%)0.75 (-0.86%)
deb6e1e — 2026-09-03 22:38:337.41 (n/a)6.45 (n/a)6.69 (n/a)5.66 (n/a)0.76 (n/a)7.40 (n/a)6.44 (n/a)6.69 (n/a)5.66 (n/a)0.76 (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)
b13d03d — 2026-09-08 22:22:2811.71 (+15.42%)9.01 (+2.85%)8.61 (+1.28%)7.84 (-4.50%)1.55 (+97.04%)11.70 (+15.42%)9.00 (+2.85%)8.60 (+1.28%)7.83 (-4.50%)1.54 (+97.04%)
deb6e1e — 2026-09-03 22:38:3310.15 (n/a)8.76 (n/a)8.50 (n/a)8.20 (n/a)0.78 (n/a)10.14 (n/a)8.75 (n/a)8.49 (n/a)8.20 (n/a)0.78 (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)
b13d03d — 2026-09-08 22:22:283.93 (n/a)3.72 (n/a)3.80 (n/a)3.42 (n/a)0.20 (n/a)3.93 (n/a)3.72 (n/a)3.80 (n/a)3.42 (n/a)0.20 (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)
b13d03d — 2026-09-08 22:22:287.58 (+1.15%)6.71 (+1.31%)7.07 (+8.37%)5.65 (-2.07%)0.82 (+24.64%)7.57 (+1.15%)6.70 (+1.31%)7.06 (+8.37%)5.65 (-2.07%)0.82 (+24.64%)
deb6e1e — 2026-09-03 22:38:337.49 (n/a)6.62 (n/a)6.52 (n/a)5.77 (n/a)0.66 (n/a)7.49 (n/a)6.62 (n/a)6.52 (n/a)5.77 (n/a)0.66 (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)
b13d03d — 2026-09-08 22:22:2811.73 (-16.52%)9.49 (-24.31%)9.40 (-31.58%)8.19 (+9.13%)1.44 (-48.75%)11.73 (-16.52%)9.49 (-24.31%)9.39 (-31.58%)8.19 (+9.13%)1.44 (-48.75%)
deb6e1e — 2026-09-03 22:38:3314.05 (n/a)12.54 (n/a)13.73 (n/a)7.51 (n/a)2.82 (n/a)14.05 (n/a)12.54 (n/a)13.72 (n/a)7.50 (n/a)2.82 (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)
b13d03d — 2026-09-08 22:22:282.93 (-6.98%)2.52 (+0.07%)2.74 (-9.84%)1.47 (+34.65%)0.59 (-32.97%)2.92 (-6.98%)2.51 (+0.07%)2.73 (-9.84%)1.46 (+34.65%)0.59 (-32.97%)
deb6e1e — 2026-09-03 22:38:333.15 (n/a)2.52 (n/a)3.04 (n/a)1.09 (n/a)0.89 (n/a)3.14 (n/a)2.51 (n/a)3.03 (n/a)1.09 (n/a)0.89 (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)
b13d03d — 2026-09-08 22:22:280.63 (+9.94%)0.40 (+8.70%)0.44 (+27.01%)0.08 (+0.02%)0.22 (+9.51%)0.62 (+9.94%)0.40 (+8.70%)0.44 (+27.01%)0.08 (+0.02%)0.21 (+9.51%)
deb6e1e — 2026-09-03 22:38:330.57 (n/a)0.37 (n/a)0.35 (n/a)0.08 (n/a)0.20 (n/a)0.56 (n/a)0.37 (n/a)0.34 (n/a)0.08 (n/a)0.20 (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)
b13d03d — 2026-09-08 22:22:280.74 (+1.81%)0.62 (+51.69%)0.66 (+42.89%)0.39 (+404.07%)0.14 (-43.49%)0.73 (+1.81%)0.61 (+51.69%)0.66 (+42.89%)0.39 (+404.07%)0.13 (-43.49%)
deb6e1e — 2026-09-03 22:38:330.73 (n/a)0.41 (n/a)0.46 (n/a)0.08 (n/a)0.24 (n/a)0.72 (n/a)0.41 (n/a)0.46 (n/a)0.08 (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)
b13d03d — 2026-09-08 22:22:282.50 (+1.55%)1.57 (-0.51%)1.68 (-5.60%)0.62 (+41.04%)0.91 (+18.50%)2.46 (+1.55%)1.55 (-0.51%)1.65 (-5.60%)0.61 (+41.04%)0.89 (+18.50%)
deb6e1e — 2026-09-03 22:38:332.47 (n/a)1.58 (n/a)1.78 (n/a)0.44 (n/a)0.76 (n/a)2.43 (n/a)1.55 (n/a)1.75 (n/a)0.43 (n/a)0.75 (n/a)

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

No metrics available.

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

No metrics available.

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

No metrics available.

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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.03 (n/a)0.03 (n/a)0.03 (n/a)0.00 (n/a)301.20 (n/a)290.46 (n/a)299.80 (n/a)255.80 (n/a)19.49 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)526.30 (n/a)430.96 (n/a)480.90 (n/a)244.30 (n/a)116.41 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)626.20 (n/a)437.90 (n/a)473.80 (n/a)243.90 (n/a)167.04 (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)
b13d03d — 2026-09-08 22:22:280.04 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)630.90 (n/a)453.86 (n/a)557.40 (n/a)233.60 (n/a)179.06 (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)
b13d03d — 2026-09-08 22:22:280.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)594.70 (n/a)380.08 (n/a)383.40 (n/a)220.30 (n/a)152.04 (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)
b13d03d — 2026-09-08 22:22:280.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)656.00 (n/a)466.52 (n/a)445.50 (n/a)304.40 (n/a)141.16 (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)
b13d03d — 2026-09-08 22:22:280.03 (-1.35%)0.03 (+29.53%)0.03 (+82.15%)0.02 (+14.88%)0.01 (-24.22%)516.00 (-12.94%)328.26 (-27.36%)285.10 (-45.10%)256.60 (+1.38%)107.84 (-33.61%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)592.70 (n/a)451.90 (n/a)519.30 (n/a)253.10 (n/a)162.42 (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)
b13d03d — 2026-09-08 22:22:280.05 (+69.10%)0.03 (+39.13%)0.03 (+66.53%)0.02 (+13.78%)0.01 (+62.45%)512.00 (-12.12%)314.92 (-26.02%)297.10 (-39.96%)153.40 (-40.86%)128.16 (-14.73%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)582.60 (n/a)425.66 (n/a)494.80 (n/a)259.40 (n/a)150.31 (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)
b13d03d — 2026-09-08 22:22:280.04 (-5.80%)0.03 (-2.66%)0.02 (-8.97%)0.02 (-3.29%)0.01 (-1.66%)502.70 (+3.39%)357.24 (+2.90%)333.90 (+9.84%)231.40 (+6.15%)123.93 (+3.08%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)486.20 (n/a)347.18 (n/a)304.00 (n/a)218.00 (n/a)120.23 (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)
b13d03d — 2026-09-08 22:22:280.03 (-5.90%)0.02 (-3.63%)0.02 (-1.05%)0.02 (+0.24%)0.01 (-12.63%)489.80 (-0.24%)424.28 (+2.77%)445.70 (+1.04%)280.80 (+6.28%)82.22 (-8.66%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)491.00 (n/a)412.84 (n/a)441.10 (n/a)264.20 (n/a)90.01 (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)
b13d03d — 2026-09-08 22:22:280.03 (-45.77%)0.02 (+1.89%)0.03 (+58.40%)0.01 (-8.48%)0.01 (-60.18%)601.30 (+9.25%)357.56 (-14.05%)290.00 (-36.87%)283.20 (+84.50%)137.34 (-11.47%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.02 (n/a)550.40 (n/a)416.02 (n/a)459.40 (n/a)153.50 (n/a)155.14 (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)
b13d03d — 2026-09-08 22:22:280.03 (+85.96%)0.02 (+23.90%)0.02 (+1.62%)0.01 (+0.86%)0.01 (+367.64%)582.10 (-0.85%)431.80 (-11.92%)465.60 (-1.59%)242.10 (-46.21%)134.19 (+140.60%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)587.10 (n/a)490.24 (n/a)473.10 (n/a)450.10 (n/a)55.77 (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)
b13d03d — 2026-09-08 22:22:280.03 (+2.69%)0.02 (+28.96%)0.02 (+3.84%)0.01 (+261.32%)0.01 (-17.19%)558.80 (-72.32%)416.02 (-45.28%)488.40 (-3.69%)243.50 (-2.64%)140.37 (-80.32%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)0.01 (n/a)2018.90 (n/a)760.24 (n/a)507.10 (n/a)250.10 (n/a)713.23 (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)
b13d03d — 2026-09-08 22:22:280.03 (+11.75%)0.02 (+10.07%)0.02 (+16.83%)0.01 (+2.44%)0.01 (+14.40%)610.30 (-2.38%)468.82 (-8.63%)459.20 (-14.41%)276.80 (-10.51%)125.78 (-4.04%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)625.20 (n/a)513.10 (n/a)536.50 (n/a)309.30 (n/a)131.08 (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)
b13d03d — 2026-09-08 22:22:280.03 (-24.03%)0.02 (-29.21%)0.02 (-32.55%)0.01 (-5.99%)0.01 (-28.48%)579.20 (+6.37%)468.86 (+37.67%)461.10 (+48.26%)292.50 (+31.64%)112.80 (-6.87%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)544.50 (n/a)340.56 (n/a)311.00 (n/a)222.20 (n/a)121.11 (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)
b13d03d — 2026-09-08 22:22:280.03 (-20.85%)0.02 (-20.48%)0.02 (-24.85%)0.02 (+12.17%)0.00 (-52.79%)480.40 (-10.86%)423.48 (+15.01%)448.30 (+33.07%)295.10 (+26.38%)76.16 (-46.56%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)538.90 (n/a)368.22 (n/a)336.90 (n/a)233.50 (n/a)142.52 (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)
b13d03d — 2026-09-08 22:22:280.03 (-23.17%)0.02 (-14.18%)0.02 (+5.82%)0.01 (-9.27%)0.01 (-30.62%)646.90 (+10.22%)456.66 (+10.53%)473.60 (-5.49%)279.60 (+30.17%)168.33 (-1.88%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)586.90 (n/a)413.14 (n/a)501.10 (n/a)214.80 (n/a)171.55 (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)
b13d03d — 2026-09-08 22:22:280.03 (-28.80%)0.02 (-40.18%)0.02 (-47.73%)0.02 (-18.50%)0.00 (-41.79%)538.20 (+22.68%)443.92 (+62.57%)447.70 (+91.32%)314.70 (+40.43%)92.19 (-0.80%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.03 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)438.70 (n/a)273.06 (n/a)234.00 (n/a)224.10 (n/a)92.93 (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)
b13d03d — 2026-09-08 22:22:280.03 (-0.95%)0.02 (-28.13%)0.01 (-47.88%)0.01 (-21.25%)0.01 (+16.45%)773.60 (+27.01%)552.90 (+46.52%)598.60 (+91.86%)244.20 (+0.95%)194.99 (+33.09%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)609.10 (n/a)377.36 (n/a)312.00 (n/a)241.90 (n/a)146.51 (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)
b13d03d — 2026-09-08 22:22:280.03 (-14.98%)0.02 (-3.36%)0.02 (-4.32%)0.01 (+200.70%)0.01 (-45.15%)607.50 (-66.75%)504.84 (-28.23%)537.30 (+4.51%)304.80 (+17.59%)118.59 (-81.43%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.00 (n/a)0.01 (n/a)1826.80 (n/a)703.40 (n/a)514.10 (n/a)259.20 (n/a)638.71 (n/a)
iron/operators/repeat

test_cols_without_a_legal_split_is_rejected[cols_1031-why_prime > 1023: the only divisors are 1 and cols, neither legal]

No metrics available.

test_cols_without_a_legal_split_is_rejected[cols_2062-why_2 x 1031: the only word-aligned chunk leaves a 1031-wide chunk count]

No metrics available.

test_cols_without_a_legal_split_is_rejected[cols_513-why_odd: every divisor is odd, so no chunk is a whole 32-bit word]

No metrics available.

test_repeat[rows_4-cols_1024-repeat_2-transfer_size_None]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.10 (-6.02%)0.07 (-17.10%)0.06 (-33.01%)0.04 (-10.81%)0.03 (+24.22%)595.90 (+12.12%)415.12 (+27.01%)439.60 (+49.27%)240.70 (+6.41%)157.93 (+33.88%)
deb6e1e — 2026-09-03 22:38:330.11 (n/a)0.08 (n/a)0.08 (n/a)0.05 (n/a)0.02 (n/a)531.50 (n/a)326.84 (n/a)294.50 (n/a)226.20 (n/a)117.97 (n/a)

test_repeat[rows_8-cols_512-repeat_4-transfer_size_64]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.16 (-8.56%)0.12 (-14.31%)0.13 (-5.92%)0.08 (-15.53%)0.03 (+6.40%)520.70 (+18.39%)363.54 (+19.18%)309.60 (+6.28%)255.20 (+9.39%)110.27 (+36.58%)
deb6e1e — 2026-09-03 22:38:330.18 (n/a)0.14 (n/a)0.14 (n/a)0.09 (n/a)0.03 (n/a)439.80 (n/a)305.04 (n/a)291.30 (n/a)233.30 (n/a)80.73 (n/a)

test_repeat[rows_8-cols_64-repeat_4-transfer_size_None]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.02 (-2.61%)0.01 (+13.93%)0.02 (+37.72%)0.01 (+11.97%)0.00 (-9.31%)517.90 (-10.68%)377.92 (-13.94%)330.70 (-27.38%)236.20 (+2.65%)127.63 (-11.15%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)579.80 (n/a)439.12 (n/a)455.40 (n/a)230.10 (n/a)143.64 (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)
b13d03d — 2026-09-08 22:22:280.03 (-24.03%)0.02 (-32.16%)0.02 (-40.36%)0.01 (-28.72%)0.01 (-32.59%)672.10 (+40.28%)471.94 (+44.80%)454.40 (+67.68%)293.70 (+31.59%)139.56 (+22.17%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)479.10 (n/a)325.92 (n/a)271.00 (n/a)223.20 (n/a)114.23 (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)
b13d03d — 2026-09-08 22:22:280.05 (+17.06%)0.04 (+4.01%)0.04 (+1.73%)0.02 (+5.48%)0.01 (+31.05%)523.30 (-5.18%)370.44 (-0.23%)292.60 (-1.68%)235.40 (-14.59%)140.67 (+16.78%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.04 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)551.90 (n/a)371.30 (n/a)297.60 (n/a)275.60 (n/a)120.46 (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)
b13d03d — 2026-09-08 22:22:280.03 (+15.04%)0.03 (+35.34%)0.03 (+46.22%)0.02 (+17.07%)0.01 (+25.69%)496.50 (-14.57%)328.50 (-25.22%)293.00 (-31.61%)244.20 (-13.07%)104.36 (-4.06%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)581.20 (n/a)439.26 (n/a)428.40 (n/a)280.90 (n/a)108.78 (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)
b13d03d — 2026-09-08 22:22:280.04 (-28.67%)0.03 (-6.57%)0.03 (+27.19%)0.02 (+40.65%)0.01 (-55.09%)534.80 (-28.90%)405.52 (-12.72%)387.00 (-21.37%)263.20 (+40.15%)107.94 (-54.50%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)0.02 (n/a)752.20 (n/a)464.64 (n/a)492.20 (n/a)187.80 (n/a)237.23 (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)
b13d03d — 2026-09-08 22:22:280.03 (+30.69%)0.02 (+6.26%)0.02 (+0.59%)0.01 (-36.38%)0.01 (+117.51%)887.50 (+57.19%)520.86 (+9.26%)489.70 (-0.59%)250.80 (-23.47%)247.08 (+166.06%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)564.60 (n/a)476.72 (n/a)492.60 (n/a)327.70 (n/a)92.86 (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)
b13d03d — 2026-09-08 22:22:280.03 (-25.01%)0.02 (-27.23%)0.02 (-47.80%)0.01 (-12.37%)0.01 (-37.89%)695.50 (+14.11%)498.98 (+27.68%)535.80 (+91.56%)297.20 (+33.33%)158.12 (-12.98%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.03 (n/a)0.04 (n/a)0.02 (n/a)0.01 (n/a)609.50 (n/a)390.82 (n/a)279.70 (n/a)222.90 (n/a)181.70 (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)
b13d03d — 2026-09-08 22:22:280.03 (-31.78%)0.02 (-31.84%)0.02 (-38.08%)0.00 (-73.97%)0.01 (-16.55%)1921.80 (+284.13%)668.42 (+109.96%)449.00 (+61.51%)239.30 (+46.63%)708.07 (+381.02%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.03 (n/a)0.03 (n/a)0.02 (n/a)0.01 (n/a)500.30 (n/a)318.36 (n/a)278.00 (n/a)163.20 (n/a)147.20 (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)
b13d03d — 2026-09-08 22:22:280.04 (+11.26%)0.02 (-3.48%)0.02 (-15.40%)0.01 (-41.66%)0.01 (+29.91%)1776.70 (+71.40%)703.92 (+31.77%)551.50 (+18.20%)218.60 (-10.12%)615.16 (+107.31%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)1036.60 (n/a)534.22 (n/a)466.60 (n/a)243.20 (n/a)296.73 (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)
b13d03d — 2026-09-08 22:22:280.04 (-27.64%)0.02 (-37.14%)0.02 (-52.28%)0.01 (-28.09%)0.01 (-24.82%)783.30 (+39.06%)514.62 (+59.04%)544.50 (+109.58%)229.00 (+38.20%)203.34 (+32.01%)
deb6e1e — 2026-09-03 22:38:330.05 (n/a)0.03 (n/a)0.03 (n/a)0.01 (n/a)0.01 (n/a)563.30 (n/a)323.58 (n/a)259.80 (n/a)165.70 (n/a)154.03 (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)
b13d03d — 2026-09-08 22:22:280.05 (+20.97%)0.02 (+24.02%)0.02 (+1.12%)0.01 (+140.89%)0.02 (+16.35%)684.90 (-58.49%)474.80 (-31.70%)526.70 (-1.11%)184.90 (-17.34%)210.98 (-61.74%)
deb6e1e — 2026-09-03 22:38:330.04 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.01 (n/a)1649.90 (n/a)695.12 (n/a)532.60 (n/a)223.70 (n/a)551.46 (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)
b13d03d — 2026-09-08 22:22:280.03 (+26.49%)0.02 (+17.55%)0.02 (-2.41%)0.02 (+17.32%)0.01 (+68.63%)539.80 (-14.76%)422.04 (-10.96%)487.60 (+2.46%)255.50 (-20.95%)130.25 (+18.36%)
deb6e1e — 2026-09-03 22:38:330.03 (n/a)0.02 (n/a)0.02 (n/a)0.01 (n/a)0.00 (n/a)633.30 (n/a)474.00 (n/a)475.90 (n/a)323.20 (n/a)110.04 (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)
b13d03d — 2026-09-08 22:22:280.41 (+16.45%)0.28 (-3.17%)0.22 (-35.18%)0.21 (+12.25%)0.09 (+24.25%)467.60 (-10.92%)379.46 (+5.06%)449.70 (+54.27%)242.00 (-14.12%)112.00 (+1.93%)
deb6e1e — 2026-09-03 22:38:330.35 (n/a)0.29 (n/a)0.34 (n/a)0.19 (n/a)0.08 (n/a)524.90 (n/a)361.20 (n/a)291.50 (n/a)281.80 (n/a)109.88 (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)
b13d03d — 2026-09-08 22:22:280.28 (-5.33%)0.22 (+2.34%)0.21 (+3.17%)0.17 (-7.70%)0.05 (+11.74%)581.80 (+8.34%)460.34 (-1.07%)469.00 (-3.08%)349.10 (+5.63%)106.66 (+24.45%)
deb6e1e — 2026-09-03 22:38:330.30 (n/a)0.22 (n/a)0.20 (n/a)0.18 (n/a)0.05 (n/a)537.00 (n/a)465.30 (n/a)483.90 (n/a)330.50 (n/a)85.71 (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)
b13d03d — 2026-09-08 22:22:280.40 (+12.11%)0.25 (+19.62%)0.18 (-8.80%)0.15 (+268.44%)0.12 (+3.19%)657.50 (-72.86%)462.12 (-43.32%)547.70 (+9.65%)244.80 (-10.82%)190.06 (-78.97%)
deb6e1e — 2026-09-03 22:38:330.36 (n/a)0.21 (n/a)0.20 (n/a)0.04 (n/a)0.12 (n/a)2422.70 (n/a)815.30 (n/a)499.50 (n/a)274.50 (n/a)903.57 (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)
b13d03d — 2026-09-08 22:22:280.25 (-16.44%)0.18 (-21.56%)0.15 (-36.47%)0.15 (+3.49%)0.04 (-35.44%)487.10 (-3.37%)434.78 (+22.97%)479.80 (+57.41%)299.70 (+19.69%)79.27 (-26.82%)
deb6e1e — 2026-09-03 22:38:330.29 (n/a)0.22 (n/a)0.24 (n/a)0.15 (n/a)0.06 (n/a)504.10 (n/a)353.58 (n/a)304.80 (n/a)250.40 (n/a)108.32 (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)
b13d03d — 2026-09-08 22:22:280.37 (+6.12%)0.22 (+1.84%)0.15 (-1.49%)0.14 (+12.34%)0.11 (-2.70%)532.40 (-10.98%)395.12 (-4.35%)495.50 (+1.52%)199.10 (-5.77%)159.53 (-12.20%)
deb6e1e — 2026-09-03 22:38:330.35 (n/a)0.22 (n/a)0.15 (n/a)0.12 (n/a)0.11 (n/a)598.10 (n/a)413.10 (n/a)488.10 (n/a)211.30 (n/a)181.71 (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)
b13d03d — 2026-09-08 22:22:280.14 (-49.89%)0.11 (-20.52%)0.12 (-9.43%)0.08 (+7.68%)0.03 (-67.51%)970.00 (-7.13%)686.36 (+7.03%)636.10 (+10.41%)517.50 (+99.58%)183.85 (-36.80%)
deb6e1e — 2026-09-03 22:38:330.28 (n/a)0.14 (n/a)0.13 (n/a)0.07 (n/a)0.08 (n/a)1044.50 (n/a)641.28 (n/a)576.10 (n/a)259.30 (n/a)290.93 (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)
b13d03d — 2026-09-08 22:22:280.49 (-0.97%)0.32 (-6.26%)0.27 (-6.62%)0.25 (+18.12%)0.10 (-27.30%)524.50 (-15.35%)434.58 (+0.22%)486.50 (+7.09%)270.20 (+0.97%)104.86 (-34.78%)
deb6e1e — 2026-09-03 22:38:330.49 (n/a)0.34 (n/a)0.29 (n/a)0.21 (n/a)0.13 (n/a)619.60 (n/a)433.62 (n/a)454.30 (n/a)267.60 (n/a)160.77 (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)
b13d03d — 2026-09-08 22:22:280.46 (-2.02%)0.35 (+2.29%)0.40 (+14.57%)0.20 (-5.87%)0.12 (+15.97%)656.90 (+6.23%)421.58 (+1.15%)330.90 (-12.71%)282.20 (+2.06%)167.48 (+23.15%)
deb6e1e — 2026-09-03 22:38:330.47 (n/a)0.34 (n/a)0.35 (n/a)0.21 (n/a)0.10 (n/a)618.40 (n/a)416.80 (n/a)379.10 (n/a)276.50 (n/a)135.99 (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)
b13d03d — 2026-09-08 22:22:280.48 (-13.85%)0.35 (-8.94%)0.40 (+5.74%)0.20 (-24.72%)0.13 (+17.83%)671.00 (+32.84%)418.56 (+17.14%)329.60 (-5.42%)275.50 (+16.05%)174.82 (+77.96%)
deb6e1e — 2026-09-03 22:38:330.55 (n/a)0.39 (n/a)0.38 (n/a)0.26 (n/a)0.11 (n/a)505.10 (n/a)357.32 (n/a)348.50 (n/a)237.40 (n/a)98.24 (n/a)
iron/operators/strided_copy

test_strided_copy[chunked_transfer]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.02 (+22.27%)0.01 (+24.03%)0.01 (+29.54%)0.01 (-3.71%)0.00 (+32.46%)508.70 (+3.84%)313.44 (-17.02%)291.00 (-22.79%)208.80 (-18.21%)115.03 (+17.37%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)489.90 (n/a)377.72 (n/a)376.90 (n/a)255.30 (n/a)98.01 (n/a)

test_strided_copy[contiguous]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.02 (+23.11%)0.01 (+45.53%)0.01 (+60.35%)0.01 (+11.48%)0.00 (+20.03%)502.90 (-10.29%)323.56 (-30.77%)302.40 (-37.64%)231.70 (-18.76%)104.72 (-4.49%)
deb6e1e — 2026-09-03 22:38:330.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)560.60 (n/a)467.36 (n/a)484.90 (n/a)285.20 (n/a)109.64 (n/a)

test_strided_copy[four_channels]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.01 (-16.10%)0.01 (-9.98%)0.01 (-11.67%)0.01 (-11.25%)0.00 (-25.66%)623.90 (+12.68%)434.52 (+8.40%)414.00 (+13.21%)299.30 (+19.20%)126.13 (-3.58%)
deb6e1e — 2026-09-03 22:38:330.02 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)553.70 (n/a)400.84 (n/a)365.70 (n/a)251.10 (n/a)130.81 (n/a)

test_strided_copy[kv_slot0]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.71 (+15.57%)0.38 (-22.68%)0.42 (-14.94%)0.07 (-78.26%)0.24 (+110.69%)1922.00 (+360.03%)641.02 (+128.30%)313.00 (+17.58%)185.40 (-13.45%)723.99 (+807.52%)
deb6e1e — 2026-09-03 22:38:330.62 (n/a)0.50 (n/a)0.50 (n/a)0.32 (n/a)0.11 (n/a)417.80 (n/a)280.78 (n/a)266.20 (n/a)214.20 (n/a)79.78 (n/a)

test_strided_copy[kv_slot5]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.54 (+4.20%)0.40 (+8.36%)0.39 (+32.67%)0.25 (-3.43%)0.12 (-0.33%)523.70 (+3.54%)354.10 (-7.88%)338.00 (-24.62%)244.70 (-4.04%)116.28 (-0.15%)
deb6e1e — 2026-09-03 22:38:330.52 (n/a)0.37 (n/a)0.29 (n/a)0.26 (n/a)0.12 (n/a)505.80 (n/a)384.38 (n/a)448.40 (n/a)255.00 (n/a)116.45 (n/a)

test_strided_copy[kv_slot5_four_channels]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.35 (-37.78%)0.27 (-41.23%)0.26 (-46.23%)0.23 (-17.94%)0.05 (-54.90%)583.40 (+21.87%)501.52 (+63.61%)510.30 (+85.97%)373.00 (+60.71%)78.40 (-20.15%)
deb6e1e — 2026-09-03 22:38:330.57 (n/a)0.46 (n/a)0.48 (n/a)0.28 (n/a)0.11 (n/a)478.70 (n/a)306.54 (n/a)274.40 (n/a)232.10 (n/a)98.18 (n/a)

test_strided_copy[kv_slot5_two_channels]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.55 (+22.14%)0.33 (-15.90%)0.27 (-37.38%)0.23 (-26.96%)0.13 (+111.53%)586.10 (+36.91%)438.12 (+28.84%)492.50 (+59.70%)238.10 (-18.12%)137.16 (+132.19%)
deb6e1e — 2026-09-03 22:38:330.45 (n/a)0.40 (n/a)0.43 (n/a)0.31 (n/a)0.06 (n/a)428.10 (n/a)340.06 (n/a)308.40 (n/a)290.80 (n/a)59.07 (n/a)

test_strided_copy[kv_slot_last]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.62 (+10.60%)0.33 (-28.81%)0.27 (-44.21%)0.21 (-19.72%)0.17 (+43.06%)644.00 (+24.56%)470.34 (+50.42%)483.20 (+79.23%)214.00 (-9.55%)161.74 (+40.31%)
deb6e1e — 2026-09-03 22:38:330.56 (n/a)0.46 (n/a)0.49 (n/a)0.26 (n/a)0.12 (n/a)517.00 (n/a)312.68 (n/a)269.60 (n/a)236.60 (n/a)115.27 (n/a)

test_strided_copy[two_channels]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.02 (+10.60%)0.01 (+13.24%)0.01 (-6.63%)0.01 (-3.64%)0.00 (+36.20%)595.10 (+3.77%)395.00 (-8.40%)422.80 (+7.12%)254.60 (-9.59%)140.19 (+17.23%)
deb6e1e — 2026-09-03 22:38:330.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)573.50 (n/a)431.22 (n/a)394.70 (n/a)281.60 (n/a)119.59 (n/a)

test_strided_copy[two_channels_chunked]

Commit/Date Bandwidth (max)Bandwidth (mean)Bandwidth (median)Bandwidth (min)Bandwidth (stddev)Latency (max)Latency (mean)Latency (median)Latency (min)Latency (stddev)
b13d03d — 2026-09-08 22:22:280.02 (+16.14%)0.01 (+15.55%)0.01 (+47.12%)0.01 (-37.57%)0.00 (+134.28%)741.60 (+60.17%)417.54 (+0.20%)295.00 (-32.03%)259.50 (-13.90%)211.01 (+217.12%)
deb6e1e — 2026-09-03 22:38:330.01 (n/a)0.01 (n/a)0.01 (n/a)0.01 (n/a)0.00 (n/a)463.00 (n/a)416.70 (n/a)434.00 (n/a)301.40 (n/a)66.54 (n/a)

test_transfer_size_not_dividing_per_channel_share_is_rejected[iter0]

No metrics available.

test_transfer_size_not_dividing_per_channel_share_is_rejected[iter1]

No metrics available.

test_transfer_size_not_dividing_per_channel_share_is_rejected[iter2]

No metrics available.

test_transfer_size_not_dividing_per_channel_share_is_rejected[iter3]

No metrics available.

test_transfer_size_not_dividing_per_channel_share_is_rejected[iter4]

No metrics available.

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)
b13d03d — 2026-09-08 22:22:280.00 (-57.14%)0.00 (-52.17%)0.00 (-60.00%)0.00 (+0.00%)0.00 (-82.18%)18461.50 (-13.79%)16907.58 (+43.42%)16958.48 (+117.96%)14052.58 (+145.91%)1784.04 (-76.21%)
deb6e1e — 2026-09-03 22:38:330.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)21415.50 (n/a)11789.16 (n/a)7780.39 (n/a)5714.62 (n/a)7500.24 (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)
b13d03d — 2026-09-08 22:22:280.00 (-15.38%)0.00 (-25.00%)0.00 (-16.67%)0.00 (-20.00%)0.00 (-22.08%)20942.81 (+17.77%)15235.10 (+21.03%)15815.40 (+6.25%)7316.49 (+20.40%)5156.48 (-4.84%)
deb6e1e — 2026-09-03 22:38:330.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)0.00 (n/a)17782.76 (n/a)12587.83 (n/a)14885.21 (n/a)6077.00 (n/a)5418.64 (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)
b13d03d — 2026-09-08 22:22:280.14 (-1.53%)0.09 (+3.36%)0.08 (+6.59%)0.08 (+6.31%)0.03 (-9.50%)27665.49 (-5.89%)23580.53 (-4.48%)24936.35 (-6.18%)14837.53 (+1.51%)5053.66 (-12.44%)
deb6e1e — 2026-09-03 22:38:330.14 (n/a)0.09 (n/a)0.08 (n/a)0.07 (n/a)0.03 (n/a)29397.45 (n/a)24686.08 (n/a)26578.49 (n/a)14617.47 (n/a)5771.46 (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)
b13d03d — 2026-09-08 22:22:281.65 (-10.62%)1.26 (-7.11%)1.61 (-2.81%)0.47 (-33.26%)0.54 (+3.59%)1109.60 (+49.84%)527.28 (+17.73%)325.00 (+2.91%)317.70 (+11.91%)342.11 (+67.13%)
deb6e1e — 2026-09-03 22:38:331.85 (n/a)1.36 (n/a)1.66 (n/a)0.71 (n/a)0.52 (n/a)740.50 (n/a)447.88 (n/a)315.80 (n/a)283.90 (n/a)204.70 (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)
b13d03d — 2026-09-08 22:22:282.60 (-15.19%)1.67 (-14.58%)1.60 (+14.96%)0.30 (-77.66%)0.92 (+13.59%)3438.10 (+347.67%)1133.26 (+87.11%)654.70 (-13.02%)403.50 (+17.91%)1296.47 (+508.66%)
deb6e1e — 2026-09-03 22:38:333.06 (n/a)1.96 (n/a)1.39 (n/a)1.37 (n/a)0.81 (n/a)768.00 (n/a)605.68 (n/a)752.70 (n/a)342.20 (n/a)213.00 (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)
b13d03d — 2026-09-08 22:22:282.69 (+37.49%)1.57 (+18.31%)1.67 (+23.40%)0.74 (-8.28%)0.82 (+93.17%)707.10 (+9.04%)429.28 (-0.36%)313.40 (-18.98%)194.80 (-27.26%)236.15 (+66.51%)
deb6e1e — 2026-09-03 22:38:331.96 (n/a)1.32 (n/a)1.36 (n/a)0.81 (n/a)0.42 (n/a)648.50 (n/a)430.84 (n/a)386.80 (n/a)267.80 (n/a)141.82 (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_09_08_22_13_55 at commit b13d03d.

Trends:

IRON Trends

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