Example Launch Social Media Posts

Bank Of America Medallion Signature Form

manhattan hotel suite design frames incredible central park view sbid

:
Merged
Product Layout Background merged 1 commit into
How To Make A Blog In WordPressfrom
Newspaper News Article Example
Sep 14, 2026
Merged

Cartoon Story On Post Its Story merged 1 commit into
Chemical Inventory Logfrom
New Food Launch Flyer

Young Child Reading Bank Of America Medallion Signature Form

Brand Story Template@jtuyls

Example Launch Social Media Posts passThrough.cc selects the wrapper element width with BIT_WIDTH. MHA already passes -DBIT_WIDTH=16 for the same kernel; mem_copy uses bfloat16 and must too. Unset BIT_WIDTH now defaults to 32. Caatering Launching Post

Child Learning To Read Bank Of America Medallion Signature Form

Product Sale Post Design Bank Of America Medallion Signature Form

Build It Credit Card Bank Of America Medallion Signature Form

Does Costco Accept Credit Cards Bank Of America Medallion Signature Form

  1. The PR is rebased on the latest devel commit and pointing to devel.
  2. Your PR has been reviewed and approved.
  3. All checks are passing.
passThrough.cc selects the wrapper element width with BIT_WIDTH. MHA already passes -DBIT_WIDTH=16 for the same kernel; mem_copy uses bfloat16 and must too. Unset BIT_WIDTH now defaults to 32.
@github-actions

Success Story Poster Template Bot commented Insta Beautey

Copy link
Copy Markdown
Contributor

Military Business Cards Bank Of America Medallion Signature Form

Young Child Reading Professional Business Card Ideas (2026_09_12_16_57_12) Things To Post On Your Story

Bank Of America Medallion Signature Form

manhattan hotel suite design frames incredible central park view sbid

:

New Post Story Insta Bubble Bank Of America Medallion Signature Form

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] - - -

Producan Instagram Post Design Bank Of America Medallion Signature Form

iron/operators/axpy
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_axpy[input_length_2048-num_aie_columns_1-tile_size_2048-scalar_factor_3.0] 197.88 439.32
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0] 143.76 439.44
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0] 184.30 780.46
test_axpy[input_length_2048-num_aie_columns_8-tile_size_256-scalar_factor_3.0] 185.78 - -
iron/operators/dequant
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32] 141.64 319.06
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32] 144.60 400.98
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32] 165.50 727.58
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32] 163.52 421.14
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32] 161.26 480.60
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32] 181.78 378.48
test_dequant[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32] 172.04 - -
test_dequant[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32] 206.04 - -
iron/operators/elementwise_add
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048] 131.86 379.68
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024] 146.10 410.02
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512] 143.60 336.10
test_elementwise_add[input_length_2048-num_aie_columns_8-tile_size_256] 159.34 - -
iron/operators/elementwise_mul
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048] 134.02 291.90
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024] 178.10 365.90
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512] 180.58 322.22
test_elementwise_mul[input_length_2048-num_aie_columns_8-tile_size_256] 170.02 - -
iron/operators/flm/gemm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_artifact_stem_differs_from_generic_gemm[M_256-K_512-N_1024] - -
test_artifact_stem_differs_from_generic_gemm[M_512-K_1024-N_2048] - -
test_gemm[M_256-K_512-N_1024-epilogue_gelu-clamp_None-rounding_conv_even] 374.90 - -
test_gemm[M_256-K_512-N_1024-epilogue_none-clamp_(-2.0, 2.0)-rounding_conv_even] 317.12 - -
test_gemm[M_256-K_512-N_1024-epilogue_none-clamp_None-rounding_conv_even] 277.44 - -
test_gemm[M_256-K_512-N_1024-epilogue_none-clamp_None-rounding_floor] 338.82 - -
test_gemm[M_256-K_512-N_1024-epilogue_silu-clamp_None-rounding_conv_even] 355.38 - -
test_gemm[M_256-K_512-N_128-epilogue_none-clamp_None-rounding_conv_even] 196.24 1081.56
test_gemm[M_256-K_512-N_1536-epilogue_none-clamp_None-rounding_conv_even] 371.70 - -
test_gemm[M_256-K_512-N_256-epilogue_none-clamp_None-rounding_conv_even] - - 1378.58
test_gemm[M_256-K_512-N_320-epilogue_none-clamp_None-rounding_conv_even] - - 948.90
test_gemm[M_256-K_512-N_512-epilogue_gelu-clamp_None-rounding_conv_even] - - 661.86
test_gemm[M_256-K_512-N_512-epilogue_none-clamp_(-2.0, 2.0)-rounding_conv_even] - - 891.40
test_gemm[M_256-K_512-N_512-epilogue_none-clamp_None-rounding_floor] - - 471.64
test_gemm[M_256-K_512-N_512-epilogue_silu-clamp_None-rounding_conv_even] - - 764.64
test_gemm[M_256-K_512-N_64-epilogue_none-clamp_None-rounding_conv_even] - - 536.08
test_gemm[M_512-K_1024-N_2048-epilogue_none-clamp_None-rounding_conv_even] 431.48 - -
test_gemm[M_512-K_1024-N_512-epilogue_none-clamp_None-rounding_conv_even] - - 1077.38
test_gemm_split_leg_windowing[iter0] - -
test_gemm_split_leg_windowing[iter1] - -
test_gemm_split_leg_windowing[iter2] - -
test_gemm_split_leg_windowing[iter3] - -
test_gemm_split_leg_windowing[iter4] - -
test_gemm_split_leg_windowing_runs[iter0] - -
test_gemm_split_leg_windowing_runs[iter1] - -
test_gemm_split_leg_windowing_runs[iter2] - -
test_gemm_split_leg_windowing_runs[iter3] - -
test_gemm_split_leg_windowing_runs[iter4] - -
test_gemm_tile_options[tn128-ma64-default] - -
test_gemm_tile_options[tn16-ma64-default] - -
test_gemm_tile_options[tn32-ma64-default] - -
test_gemm_tile_options[tn64-ma16-default] - - -
test_gemm_tile_options[tn64-ma32-default] - - -
iron/operators/gelu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_gelu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] 144.92 417.44
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] 181.72 354.64
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] 165.38 441.00
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] 175.40 365.50
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] 168.66 412.52
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] 158.46 426.92
test_gelu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] 171.54 - -
test_gelu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] 192.38 - -
iron/operators/gemm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_gemm[M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1] 2230.72 - -
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] 282.14 536.56
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] 261.66 557.34
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_1-b_col_maj_False-c_col_maj_False-m_64-k_64-n_64-trace_size_0-partition_N_1] 47879.40 83032.92
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] 27857.74 22511.00
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1] 7951.02 - -
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] 2147.14 4402.62
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] 3463.80 6846.56
test_gemm[M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1] 1578.78 - -
iron/operators/gemv
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_gemv[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128] 0.20 0.08
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048] 12.60 3.61
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024] 24.32 6.33
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512] 39.76 10.35
test_gemv[M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256] 43.07 - -
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024] 12.60 3.66
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024] 24.26 6.27
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024] 38.91 11.44
test_gemv[M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024] 42.34 - -
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2] 8.18 1.75
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2] 0.84 0.15
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4] 1.01 0.40
test_gemv_batched[M_256-K_128-num_aie_columns_8-tile_size_input_1-tile_size_output_32-num_batches_100] 15.08 - -
test_gemv_batched[M_448-K_64-num_aie_columns_8-tile_size_input_1-tile_size_output_56-num_batches_192] 13.20 - -
test_gemv_batched[M_512-K_64-num_aie_columns_8-tile_size_input_4-tile_size_output_64-num_batches_32] 6.89 - -
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8] 5.26 0.98
test_gemv_gelu[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128] 0.16 -
test_gemv_gelu[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048] 11.65 -
test_gemv_gelu[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024] 12.19 -
iron/operators/layer_norm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] 149.72 385.80
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] 168.56 377.86
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] 203.98 276.34
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] 158.02 431.50
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] 187.80 312.68
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] 187.94 428.78
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] 173.46 - -
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] 213.08 - -
iron/operators/leaky_relu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.01] 162.24 326.32
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1] 172.62 336.06
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25] 181.04 267.64
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01] 164.00 337.94
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01] 157.62 338.62
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01] 176.92 391.26
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01] 164.86 380.66
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01] 189.84 522.74
test_leaky_relu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-alpha_0.01] 183.18 - -
test_leaky_relu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-alpha_0.01] 233.74 - -
iron/operators/mem_copy
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048] 167.84 316.38
test_mem_copy[input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128] 218.34 - -
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024] 167.82 387.94
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024] 164.36 452.36
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512] 164.70 327.28
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512] 217.22 389.74
test_mem_copy[input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256] 215.62 - -
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256] 177.86 480.66
iron/operators/mha
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_mha[seq_len_16384-dim_64-num_heads_1-num_pipelines_8-num_kv_heads_0] 47536.68 - -
iron/operators/relu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] 183.34 281.10
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] 170.24 652.22
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] 189.56 392.26
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] 172.34 338.42
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] 162.92 414.56
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] 196.34 437.40
test_relu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] 186.74 - -
test_relu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] 222.42 - -
iron/operators/repeat
Test Krackan Status Krackan Latency (mean) 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] 157.26 444.52
test_repeat[rows_8-cols_512-repeat_4-transfer_size_64] 177.26 386.22
test_repeat[rows_8-cols_64-repeat_4-transfer_size_None] 166.96 363.28
iron/operators/rms_norm
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_False] 154.12 700.64
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True] 151.72 409.98
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False] 174.78 446.52
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True] 170.96 434.24
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False] 176.18 521.18
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True] 179.18 463.80
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False] 161.52 354.40
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True] 203.02 415.66
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False] 164.36 682.14
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True] 172.00 495.82
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False] 190.52 492.80
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_True] 208.00 - -
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False] 163.28 - -
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_True] 228.98 - -
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False] 203.68 - -
iron/operators/rope
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0] 155.52 454.64
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0] 168.74 536.38
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0] 160.38 517.64
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0] 223.92 - -
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0] 162.12 397.34
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0] 153.74 414.68
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0] 162.16 436.98
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0] 178.28 - -
iron/operators/sigmoid
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] 159.44 425.18
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] 151.02 469.64
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] 151.64 385.46
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] 167.52 407.72
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] 158.78 538.32
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] 193.62 461.20
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] 193.54 - -
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] 191.76 - -
iron/operators/silu
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_silu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] 139.60 381.44
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] 148.70 420.52
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] 137.90 471.88
test_silu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] 170.58 - -
iron/operators/softmax
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024] 179.10 499.20
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048] 170.96 419.54
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512] 170.82 776.36
iron/operators/strided_copy
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_strided_copy[chunked_transfer] 160.06 309.72
test_strided_copy[contiguous] 179.52 403.94
test_strided_copy[four_channels] 147.98 331.46
test_strided_copy[kv_slot0] 150.36 295.74
test_strided_copy[kv_slot5] 167.68 308.62
test_strided_copy[kv_slot5_four_channels] 213.02 356.24
test_strided_copy[kv_slot5_two_channels] 175.40 362.02
test_strided_copy[kv_slot_last] 167.84 264.26
test_strided_copy[two_channels] 138.58 256.76
test_strided_copy[two_channels_chunked] 152.74 317.20
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 Latency (mean) Phoenix Status Phoenix Latency (mean)
test_swiglu_decode[embedding_dim_1024-hidden_dim_3584] 967.12 15432.86
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048] 1061.22 15124.63
iron/operators/swiglu_prefill
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_swiglu_prefill[seq_len_256-embedding_dim_2048-hidden_dim_2048-prio_accuracy_False] 2203.45 22517.06
iron/operators/tanh
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_tanh[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048] 150.20 419.90
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024] 165.12 515.56
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024] 164.58 345.74
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512] 181.00 455.70
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512] 187.74 342.90
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256] 171.96 405.76
test_tanh[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256] 149.92 - -
test_tanh[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128] 194.10 - -
iron/operators/transpose
Test Krackan Status Krackan Latency (mean) Phoenix Status Phoenix Latency (mean)
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_1] 192.38 857.28
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2] 247.50 630.48
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1] 175.00 503.14
Krackan - Small

Bank Of America Medallion Signature Form

manhattan hotel suite design frames incredible central park view sbid

:

Days To Go Post Ideas Tested on 2026_09_12_16_57_12 at commit f09f333. What Is Apr For Credit Cards

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]✅ 5/5197.880.07n/a
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]✅ 5/5143.760.09n/a
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]✅ 5/5184.300.07n/a
test_axpy[input_length_2048-num_aie_columns_8-tile_size_256-scalar_factor_3.0]✅ 5/5185.780.07n/a
iron/operators/dequant
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_dequant[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-group_size_32]✅ 5/5141.640.04n/a
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]✅ 5/5144.600.04n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]✅ 5/5165.500.03n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]✅ 5/5163.520.03n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]✅ 5/5161.260.03n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]✅ 5/5181.780.03n/a
test_dequant[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-group_size_32]✅ 5/5172.040.03n/a
test_dequant[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-group_size_32]✅ 5/5206.040.03n/a
iron/operators/elementwise_add
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5131.860.09n/a
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5146.100.09n/a
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5143.600.09n/a
test_elementwise_add[input_length_2048-num_aie_columns_8-tile_size_256]✅ 5/5159.340.08n/a
iron/operators/elementwise_mul
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5134.020.09n/a
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5178.100.07n/a
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5180.580.07n/a
test_elementwise_mul[input_length_2048-num_aie_columns_8-tile_size_256]✅ 5/5170.020.07n/a
iron/operators/flm/gemm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_artifact_stem_differs_from_generic_gemm[M_256-K_512-N_1024]✅ 5/5n/an/an/a
test_artifact_stem_differs_from_generic_gemm[M_512-K_1024-N_2048]✅ 5/5n/an/an/a
test_gemm[M_256-K_512-N_1024-epilogue_gelu-clamp_None-rounding_conv_even]✅ 5/5374.903.91762.90
test_gemm[M_256-K_512-N_1024-epilogue_none-clamp_(-2.0, 2.0)-rounding_conv_even]✅ 5/5317.124.76929.32
test_gemm[M_256-K_512-N_1024-epilogue_none-clamp_None-rounding_conv_even]✅ 5/5277.445.481069.30
test_gemm[M_256-K_512-N_1024-epilogue_none-clamp_None-rounding_floor]✅ 5/5338.824.69914.49
test_gemm[M_256-K_512-N_1024-epilogue_silu-clamp_None-rounding_conv_even]✅ 5/5355.384.02783.98
test_gemm[M_256-K_512-N_128-epilogue_none-clamp_None-rounding_conv_even]✅ 5/5196.242.09174.39
test_gemm[M_256-K_512-N_1536-epilogue_none-clamp_None-rounding_conv_even]✅ 5/5371.705.221087.83
test_gemm[M_512-K_1024-N_2048-epilogue_none-clamp_None-rounding_conv_even]✅ 5/5431.4813.115115.86
test_gemm_split_leg_windowing[iter0]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter1]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter2]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter3]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter4]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter0]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter1]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter2]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter3]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter4]✅ 1/1n/an/an/a
test_gemm_tile_options[tn128-ma64-default]✅ 5/5n/an/an/a
test_gemm_tile_options[tn16-ma64-default]✅ 5/5n/an/an/a
test_gemm_tile_options[tn32-ma64-default]✅ 5/5n/an/an/a
test_gemm_tile_options[tn64-ma32-default]✅ 5/5n/an/an/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]✅ 5/5144.920.06n/a
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5181.720.05n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5165.380.05n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5175.400.05n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5168.660.05n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5158.460.05n/a
test_gelu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5171.540.05n/a
test_gelu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5192.380.04n/a
iron/operators/gemm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_gemm[M_1792-K_896-N_1152-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_64-k_32-n_48-trace_size_0-partition_N_1]✅ 5/52230.724.221658.74
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]✅ 5/5282.140.8134.74
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]✅ 5/5261.660.9440.13
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]✅ 5/547879.400.53358.82
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]✅ 5/527857.740.90616.71
test_gemm[M_2048-K_2048-N_2048-num_aie_columns_8-b_col_maj_True-c_col_maj_True-m_64-k_64-n_64-trace_size_0-partition_N_1]✅ 5/57951.023.172161.47
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]✅ 5/52147.143.831003.87
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]✅ 5/53463.800.3719.80
test_gemm[M_896-K_1792-N_640-num_aie_columns_8-b_col_maj_False-c_col_maj_True-m_32-k_64-n_80-trace_size_0-partition_N_1]✅ 5/51578.784.271319.04
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/a0.200.20
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a12.6012.60
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]✅ 5/5n/a24.3224.31
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]✅ 5/5n/a39.7639.73
test_gemv[M_2048-K_8192-num_aie_columns_8-tile_size_input_1-tile_size_output_256]✅ 5/5n/a43.0743.04
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a12.6012.60
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a24.2624.25
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a38.9138.88
test_gemv[M_8192-K_2048-num_aie_columns_8-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a42.3442.31
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2]✅ 5/5n/a8.188.16
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2]✅ 5/5n/a0.840.83
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4]✅ 5/5n/a1.011.00
test_gemv_batched[M_256-K_128-num_aie_columns_8-tile_size_input_1-tile_size_output_32-num_batches_100]✅ 5/5n/a15.0814.90
test_gemv_batched[M_448-K_64-num_aie_columns_8-tile_size_input_1-tile_size_output_56-num_batches_192]✅ 5/5n/a13.2012.97
test_gemv_batched[M_512-K_64-num_aie_columns_8-tile_size_input_4-tile_size_output_64-num_batches_32]✅ 5/5n/a6.896.77
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8]✅ 5/5n/a5.265.17
test_gemv_gelu[M_128-K_128-num_aie_columns_1-tile_size_input_32-tile_size_output_128]✅ 5/5n/a0.160.16
test_gemv_gelu[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a11.6511.65
test_gemv_gelu[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a12.1912.18
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]✅ 5/5149.720.06n/a
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5168.560.05n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5203.980.04n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5158.020.05n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5187.800.04n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5187.940.05n/a
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5173.460.05n/a
test_layer_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5213.080.04n/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]✅ 5/5162.240.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1]✅ 5/5172.620.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25]✅ 5/5181.040.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01]✅ 5/5164.000.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01]✅ 5/5157.620.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01]✅ 5/5176.920.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01]✅ 5/5164.860.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01]✅ 5/5189.840.04n/a
test_leaky_relu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-alpha_0.01]✅ 5/5183.180.05n/a
test_leaky_relu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-alpha_0.01]✅ 5/5233.740.04n/a
iron/operators/mem_copy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_mem_copy[input_length_2048-num_cores_1-num_channels_1-bypass_False-tile_size_2048]✅ 5/5167.840.05n/a
test_mem_copy[input_length_2048-num_cores_16-num_channels_2-bypass_False-tile_size_128]✅ 5/5218.340.04n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]✅ 5/5167.820.05n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]✅ 5/5164.360.05n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]✅ 5/5164.700.05n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]✅ 5/5217.220.04n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_1-bypass_False-tile_size_256]✅ 5/5215.620.04n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]✅ 5/5177.860.05n/a
iron/operators/mha
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_mha[seq_len_16384-dim_64-num_heads_1-num_pipelines_8-num_kv_heads_0]✅ 5/547536.680.18n/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]✅ 5/5183.340.05n/a
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5170.240.05n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5189.560.04n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5172.340.05n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5162.920.05n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5196.340.04n/a
test_relu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5186.740.04n/a
test_relu[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5222.420.04n/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]✅ 5/5157.260.16n/a
test_repeat[rows_8-cols_512-repeat_4-transfer_size_64]✅ 5/5177.260.24n/a
test_repeat[rows_8-cols_64-repeat_4-transfer_size_None]✅ 5/5166.960.03n/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]✅ 5/5154.120.06n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]✅ 5/5151.720.08n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]✅ 5/5174.780.05n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]✅ 5/5170.960.06n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]✅ 5/5176.180.05n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]✅ 5/5179.180.06n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]✅ 5/5161.520.05n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]✅ 5/5203.020.05n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]✅ 5/5164.360.05n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]✅ 5/5172.000.06n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]✅ 5/5190.520.05n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_True]✅ 5/5208.000.04n/a
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_False]✅ 5/5163.280.05n/a
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256-weighted_True]✅ 5/5228.980.04n/a
test_rms_norm[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128-weighted_False]✅ 5/5203.680.04n/a
iron/operators/rope
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0]✅ 5/5155.520.64n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]✅ 5/5168.740.60n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]✅ 5/5160.380.63n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_8-method_type_0]✅ 5/5223.920.47n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]✅ 5/5162.120.46n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]✅ 5/5153.740.51n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]✅ 5/5162.160.46n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_8-method_type_0]✅ 5/5178.280.42n/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]✅ 5/5159.440.05n/a
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5151.020.06n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5151.640.06n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5167.520.05n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5158.780.05n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5193.620.05n/a
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5193.540.04n/a
test_sigmoid[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5191.760.04n/a
iron/operators/silu
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_silu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5139.600.06n/a
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5148.700.06n/a
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5137.900.06n/a
test_silu[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5170.580.05n/a
iron/operators/softmax
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_1024]✅ 5/5179.100.74n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]✅ 5/5170.960.77n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5170.820.80n/a
iron/operators/strided_copy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_strided_copy[chunked_transfer]✅ 5/5160.060.03n/a
test_strided_copy[contiguous]✅ 5/5179.520.02n/a
test_strided_copy[four_channels]✅ 5/5147.980.03n/a
test_strided_copy[kv_slot0]✅ 5/5150.360.91n/a
test_strided_copy[kv_slot5]✅ 5/5167.680.83n/a
test_strided_copy[kv_slot5_four_channels]✅ 5/5213.020.70n/a
test_strided_copy[kv_slot5_two_channels]✅ 5/5175.400.79n/a
test_strided_copy[kv_slot_last]✅ 5/5167.840.80n/a
test_strided_copy[two_channels]✅ 5/5138.580.03n/a
test_strided_copy[two_channels_chunked]✅ 5/5152.740.03n/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]✅ 5/5967.120.00n/a
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]✅ 5/51061.220.01n/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]✅ 5/52203.450.95n/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]✅ 5/5150.200.05n/a
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5165.120.05n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5164.580.05n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5181.000.05n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5187.740.04n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5171.960.05n/a
test_tanh[input_length_2048-num_aie_columns_8-num_channels_1-tile_size_256]✅ 5/5149.920.06n/a
test_tanh[input_length_2048-num_aie_columns_8-num_channels_2-tile_size_128]✅ 5/5194.100.04n/a
iron/operators/transpose
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_1]✅ 5/5192.382.73n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2]✅ 5/5247.504.27n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1]✅ 5/5175.003.16n/a
Krackan - Examples

Bank Of America Medallion Signature Form

manhattan hotel suite design frames incredible central park view sbid

:

Child Learning To Read Tested on 2026_09_12_17_04_17 at commit f09f333. Process Flow Chart Template

iron/applications/llama_3.2_1b
TestChecksTTFT (mean)TPS (mean)
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_1]✅ 5/52.07n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_1024_tokens_40]✅ 5/52.097.81
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_1]✅ 5/52.03n/a
test_llama_3_2_1b[llama_3.2_1b_prompt_13_tokens_40]✅ 5/52.047.59
Phoenix - Small

Bank Of America Medallion Signature Form

manhattan hotel suite design frames incredible central park view sbid

:

Product Sale Post Design Tested on 2026_09_12_16_55_58 at commit f09f333. If You Can Read This Jumbled Letters

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]✅ 5/5439.320.03n/a
test_axpy[input_length_2048-num_aie_columns_2-tile_size_1024-scalar_factor_3.0]✅ 5/5439.440.03n/a
test_axpy[input_length_2048-num_aie_columns_4-tile_size_512-scalar_factor_3.0]✅ 5/5780.460.02n/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]✅ 5/5319.060.02n/a
test_dequant[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-group_size_32]✅ 5/5400.980.01n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-group_size_32]✅ 5/5727.580.01n/a
test_dequant[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-group_size_32]✅ 5/5421.140.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-group_size_32]✅ 5/5480.600.01n/a
test_dequant[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-group_size_32]✅ 5/5378.480.02n/a
iron/operators/elementwise_add
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_add[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5379.680.04n/a
test_elementwise_add[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5410.020.03n/a
test_elementwise_add[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5336.100.04n/a
iron/operators/elementwise_mul
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_elementwise_mul[input_length_2048-num_aie_columns_1-tile_size_2048]✅ 5/5291.900.05n/a
test_elementwise_mul[input_length_2048-num_aie_columns_2-tile_size_1024]✅ 5/5365.900.04n/a
test_elementwise_mul[input_length_2048-num_aie_columns_4-tile_size_512]✅ 5/5322.220.04n/a
iron/operators/flm/gemm
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_artifact_stem_differs_from_generic_gemm[M_256-K_512-N_1024]✅ 5/5n/an/an/a
test_artifact_stem_differs_from_generic_gemm[M_512-K_1024-N_2048]✅ 5/5n/an/an/a
test_gemm[M_256-K_512-N_128-epilogue_none-clamp_None-rounding_conv_even]✅ 5/51081.560.6345.79
test_gemm[M_256-K_512-N_256-epilogue_none-clamp_None-rounding_conv_even]✅ 5/51378.580.6869.91
test_gemm[M_256-K_512-N_320-epilogue_none-clamp_None-rounding_conv_even]✅ 5/5948.901.08119.72
test_gemm[M_256-K_512-N_512-epilogue_gelu-clamp_None-rounding_conv_even]✅ 5/5661.861.62207.18
test_gemm[M_256-K_512-N_512-epilogue_none-clamp_(-2.0, 2.0)-rounding_conv_even]✅ 5/5891.401.47187.76
test_gemm[M_256-K_512-N_512-epilogue_none-clamp_None-rounding_floor]✅ 5/5471.642.27290.46
test_gemm[M_256-K_512-N_512-epilogue_silu-clamp_None-rounding_conv_even]✅ 5/5764.641.46187.14
test_gemm[M_256-K_512-N_64-epilogue_none-clamp_None-rounding_conv_even]✅ 5/5536.080.7133.05
test_gemm[M_512-K_1024-N_512-epilogue_none-clamp_None-rounding_conv_even]✅ 5/51077.382.96605.20
test_gemm_split_leg_windowing[iter0]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter1]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter2]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter3]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing[iter4]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter0]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter1]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter2]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter3]✅ 1/1n/an/an/a
test_gemm_split_leg_windowing_runs[iter4]✅ 1/1n/an/an/a
test_gemm_tile_options[tn128-ma64-default]✅ 5/5n/an/an/a
test_gemm_tile_options[tn16-ma64-default]✅ 5/5n/an/an/a
test_gemm_tile_options[tn32-ma64-default]✅ 5/5n/an/an/a
test_gemm_tile_options[tn64-ma16-default]✅ 5/5n/an/an/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]✅ 5/5417.440.02n/a
test_gelu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5354.640.03n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5441.000.02n/a
test_gelu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5365.500.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5412.520.02n/a
test_gelu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5426.920.02n/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]✅ 5/5536.560.4217.88
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]✅ 5/5557.340.4017.25
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]✅ 5/583032.920.30206.99
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]✅ 5/522511.001.12763.95
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]✅ 5/54402.622.05537.36
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]✅ 5/56846.560.1910.30
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/a0.080.08
test_gemv[M_2048-K_8192-num_aie_columns_1-tile_size_input_1-tile_size_output_2048]✅ 5/5n/a3.613.61
test_gemv[M_2048-K_8192-num_aie_columns_2-tile_size_input_1-tile_size_output_1024]✅ 5/5n/a6.336.33
test_gemv[M_2048-K_8192-num_aie_columns_4-tile_size_input_1-tile_size_output_512]✅ 5/5n/a10.3510.35
test_gemv[M_8192-K_2048-num_aie_columns_1-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a3.663.66
test_gemv[M_8192-K_2048-num_aie_columns_2-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a6.276.27
test_gemv[M_8192-K_2048-num_aie_columns_4-tile_size_input_4-tile_size_output_1024]✅ 5/5n/a11.4411.43
test_gemv_batched[M_1024-K_1024-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_2]✅ 5/5n/a1.751.74
test_gemv_batched[M_1026-K_64-num_aie_columns_1-tile_size_input_1-tile_size_output_2-num_batches_2]✅ 5/5n/a0.150.14
test_gemv_batched[M_256-K_128-num_aie_columns_1-tile_size_input_1-tile_size_output_256-num_batches_4]✅ 5/5n/a0.400.39
test_gemv_batched[M_64-K_1536-num_aie_columns_1-tile_size_input_1-tile_size_output_64-num_batches_8]✅ 5/5n/a0.980.97
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]✅ 5/5385.800.03n/a
test_layer_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5377.860.03n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5276.340.03n/a
test_layer_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5431.500.02n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5312.680.03n/a
test_layer_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5428.780.02n/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]✅ 5/5326.320.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.1]✅ 5/5336.060.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-alpha_0.25]✅ 5/5267.640.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-alpha_0.01]✅ 5/5337.940.03n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-alpha_0.01]✅ 5/5338.620.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-alpha_0.01]✅ 5/5391.260.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-alpha_0.01]✅ 5/5380.660.02n/a
test_leaky_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-alpha_0.01]✅ 5/5522.740.02n/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]✅ 5/5316.380.03n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_1-bypass_False-tile_size_1024]✅ 5/5387.940.02n/a
test_mem_copy[input_length_2048-num_cores_2-num_channels_2-bypass_False-tile_size_1024]✅ 5/5452.360.02n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_1-bypass_False-tile_size_512]✅ 5/5327.280.03n/a
test_mem_copy[input_length_2048-num_cores_4-num_channels_2-bypass_False-tile_size_512]✅ 5/5389.740.03n/a
test_mem_copy[input_length_2048-num_cores_8-num_channels_2-bypass_False-tile_size_256]✅ 5/5480.660.02n/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]✅ 5/5281.100.03n/a
test_relu[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5652.220.03n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5392.260.02n/a
test_relu[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5338.420.03n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5414.560.02n/a
test_relu[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5437.400.02n/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]✅ 5/5444.520.06n/a
test_repeat[rows_8-cols_512-repeat_4-transfer_size_64]✅ 5/5386.220.11n/a
test_repeat[rows_8-cols_64-repeat_4-transfer_size_None]✅ 5/5363.280.02n/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]✅ 5/5700.640.02n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048-weighted_True]✅ 5/5409.980.03n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_False]✅ 5/5446.520.02n/a
test_rms_norm[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024-weighted_True]✅ 5/5434.240.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_False]✅ 5/5521.180.02n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024-weighted_True]✅ 5/5463.800.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_False]✅ 5/5354.400.03n/a
test_rms_norm[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512-weighted_True]✅ 5/5415.660.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_False]✅ 5/5682.140.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512-weighted_True]✅ 5/5495.820.02n/a
test_rms_norm[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256-weighted_False]✅ 5/5492.800.02n/a
iron/operators/rope
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_1-method_type_0]✅ 5/5454.640.23n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_2-method_type_0]✅ 5/5536.380.19n/a
test_rope[rows_32-cols_512-angle_rows_32-aie_columns_4-method_type_0]✅ 5/5517.640.22n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_1-method_type_0]✅ 5/5397.340.21n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_2-method_type_0]✅ 5/5414.680.20n/a
test_rope[rows_32-cols_512-angle_rows_8-aie_columns_4-method_type_0]✅ 5/5436.980.18n/a
iron/operators/sigmoid
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_1-tile_size_2048]✅ 5/5425.180.02n/a
test_sigmoid[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5469.640.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5385.460.02n/a
test_sigmoid[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5407.720.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5538.320.02n/a
test_sigmoid[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5461.200.02n/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]✅ 5/5381.440.03n/a
test_silu[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5420.520.02n/a
test_silu[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5471.880.02n/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]✅ 5/5499.200.29n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_2048]✅ 5/5419.540.34n/a
test_softmax[input_length_32768-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5776.360.26n/a
iron/operators/strided_copy
TestChecksLatency (mean)Bandwidth (mean)Throughput (mean)
test_strided_copy[chunked_transfer]✅ 5/5309.720.01n/a
test_strided_copy[contiguous]✅ 5/5403.940.01n/a
test_strided_copy[four_channels]✅ 5/5331.460.01n/a
test_strided_copy[kv_slot0]✅ 5/5295.740.49n/a
test_strided_copy[kv_slot5]✅ 5/5308.620.46n/a
test_strided_copy[kv_slot5_four_channels]✅ 5/5356.240.39n/a
test_strided_copy[kv_slot5_two_channels]✅ 5/5362.020.42n/a
test_strided_copy[kv_slot_last]✅ 5/5264.260.50n/a
test_strided_copy[two_channels]✅ 5/5256.760.02n/a
test_strided_copy[two_channels_chunked]✅ 5/5317.200.01n/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]✅ 5/515432.860.00n/a
test_swiglu_decode[embedding_dim_2048-hidden_dim_2048]✅ 5/515124.630.00n/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]✅ 5/522517.060.10n/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]✅ 5/5419.900.02n/a
test_tanh[input_length_2048-num_aie_columns_1-num_channels_2-tile_size_1024]✅ 5/5515.560.02n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_1-tile_size_1024]✅ 5/5345.740.03n/a
test_tanh[input_length_2048-num_aie_columns_2-num_channels_2-tile_size_512]✅ 5/5455.700.02n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_1-tile_size_512]✅ 5/5342.900.03n/a
test_tanh[input_length_2048-num_aie_columns_4-num_channels_2-tile_size_256]✅ 5/5405.760.02n/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]✅ 5/5857.280.96n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_1-m_64-n_64-s_8-num_batches_2]✅ 5/5630.481.79n/a
test_transpose[M_2048-N_64-aie_columns_1-channels_2-m_64-n_64-s_8-num_batches_1]✅ 5/5503.141.23n/a
Phoenix - Examples

Bank Of America Medallion Signature Form

manhattan hotel suite design frames incredible central park view sbid

:

Build It Credit Card Tested on 2026_09_12_16_49_09 at commit f09f333. YouTube Books Read Aloud

Does Costco Accept Credit Cards Trend tables omitted, the comment hit CloneAGC's size limit. Full report in the If You Can Read This 01001. How To Write Newspaper Articles For Kids

Merged via the queue into amd:devel with commit Article Business Cards Free Shipping
5 checks passed
New Product Launching Unveiling Ideas to join this conversation on CloneAGC. Already have an account? LinkedIn Post On Phone

Facebook. Popular Creative Image Bank Of America Medallion Signature Form

None yet

Product Launch Investment Template Bank Of America Medallion Signature Form