Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
70 changes: 35 additions & 35 deletions .github/configs/nvidia-master.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -615,7 +615,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx1_gen6_tp8_batch8_eplb0_mtp3_8.yaml
Expand All @@ -630,7 +630,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx1_gen2_tp8_batch32_eplb0_mtp3_8.yaml
Expand All @@ -645,7 +645,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx1_gen6_tp8_batch8_eplb0_mtp3_48.yaml
Expand All @@ -660,7 +660,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx1_gen4_tp8_batch16_eplb0_mtp3_64.yaml
Expand All @@ -676,7 +676,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 2
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx2_gen3_dep8_batch8_eplb0_mtp3_224.yaml
Expand All @@ -691,7 +691,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 2
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx2_gen1_dep8_batch32_eplb0_mtp3_288.yaml
Expand All @@ -706,7 +706,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 4
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/mtp/ctx4_gen1_dep8_batch128_eplb0_mtp2_1088.yaml
Expand All @@ -722,8 +722,8 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
dp-attn: true
ep: 1
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen1_tp8_batch1_eplb0_mtp0_1.yaml
- "CONFIG_FILE=recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen1_tp8_batch1_eplb0_mtp0_1.yaml"
Expand All @@ -736,7 +736,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen4_tp8_batch32_eplb0_mtp0_32.yaml
Expand All @@ -750,7 +750,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen4_tp8_batch32_eplb0_mtp0_128.yaml
Expand All @@ -764,7 +764,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen6_tp8_batch16_eplb0_mtp0_96.yaml
Expand All @@ -779,7 +779,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen1_dep8_batch128_eplb0_mtp0_128.yaml
Expand All @@ -793,7 +793,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen2_dep8_batch64_eplb0_mtp0_128.yaml
Expand All @@ -807,7 +807,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 1
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx1_gen1_dep8_batch256_eplb0_mtp0_256.yaml
Expand All @@ -821,7 +821,7 @@ dsr1-fp8-b200-dynamo-trt:
prefill:
num-worker: 2
tp: 8
ep: 8
ep: 1
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/b200-fp8/8k1k/stp/ctx2_gen1_dep8_batch640_eplb0_mtp0_640.yaml
Expand Down Expand Up @@ -2111,7 +2111,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/mtp/c4_ctx1_gen11_tep8_batch128_eplb0_mtp3.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/mtp/c4_ctx1_gen11_tep8_batch128_eplb0_mtp3.yaml"
Expand All @@ -2126,7 +2126,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/mtp/c8_ctx1_gen11_tep8_batch128_eplb0_mtp3.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/mtp/c8_ctx1_gen11_tep8_batch128_eplb0_mtp3.yaml"
Expand All @@ -2141,7 +2141,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/mtp/c16_ctx1_gen9_tep8_batch128_eplb0_mtp3.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/mtp/c16_ctx1_gen9_tep8_batch128_eplb0_mtp3.yaml"
Expand All @@ -2156,7 +2156,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/mtp/c32_ctx1_gen11_tep8_batch128_eplb0_mtp3.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/mtp/c32_ctx1_gen11_tep8_batch128_eplb0_mtp3.yaml"
Expand Down Expand Up @@ -2245,7 +2245,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/stp/c4_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/stp/c4_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml"
Expand All @@ -2259,7 +2259,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/stp/c8_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/stp/c8_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml"
Expand All @@ -2273,7 +2273,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/stp/c16_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/stp/c16_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml"
Expand All @@ -2287,7 +2287,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/stp/c32_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/stp/c32_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml"
Expand All @@ -2301,7 +2301,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: false
dp-attn: true
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/1k1k/stp/c64_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/1k1k/stp/c64_ctx1_gen9_tep8_batch256_eplb0_mtp0.yaml"
Expand Down Expand Up @@ -2437,7 +2437,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/mtp/c64_ctx1_gen1_dep8_batch32_eplb0_mtp2.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/mtp/c64_ctx1_gen1_dep8_batch32_eplb0_mtp2.yaml"
Expand All @@ -2452,7 +2452,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 2
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/mtp/c128_ctx2_gen1_dep8_batch32_eplb0_mtp2.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/mtp/c128_ctx2_gen1_dep8_batch32_eplb0_mtp2.yaml"
Expand All @@ -2467,7 +2467,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 3
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/mtp/c256_ctx3_gen1_dep8_batch32_eplb0_mtp2.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/mtp/c256_ctx3_gen1_dep8_batch32_eplb0_mtp2.yaml"
Expand All @@ -2482,7 +2482,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 3
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/mtp/c512_ctx3_gen1_dep8_batch64_eplb0_mtp1.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/mtp/c512_ctx3_gen1_dep8_batch64_eplb0_mtp1.yaml"
Expand Down Expand Up @@ -2567,7 +2567,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 2
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/stp/c64_ctx2_gen3_dep8_batch128_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/stp/c64_ctx2_gen3_dep8_batch128_eplb0_mtp0.yaml"
Expand All @@ -2581,7 +2581,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 1
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/stp/c128_ctx1_gen1_dep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/stp/c128_ctx1_gen1_dep8_batch256_eplb0_mtp0.yaml"
Expand All @@ -2595,7 +2595,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 5
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/stp/c256_ctx5_gen3_dep8_batch256_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/stp/c256_ctx5_gen3_dep8_batch256_eplb0_mtp0.yaml"
Expand All @@ -2609,7 +2609,7 @@ dsr1-fp8-h200-dynamo-trt:
num-worker: 3
tp: 8
ep: 8
dp-attn: true
dp-attn: false
additional-settings:
# https://github.com/ishandhanani/srt-slurm/blob/sa-submission-q1-2026/recipes/trtllm/h200/8k1k/stp/c512_ctx3_gen1_dep8_batch512_eplb0_mtp0.yaml
- "CONFIG_FILE=recipes/trtllm/h200/8k1k/stp/c512_ctx3_gen1_dep8_batch512_eplb0_mtp0.yaml"
Expand Down Expand Up @@ -3982,7 +3982,7 @@ dsr1-fp4-gb200-dynamo-trt:
num-worker: 4
tp: 8
ep: 8
dp-attn: true
dp-attn: false
- conc-list: [ 5 ]
prefill:
num-worker: 1
Expand All @@ -3996,7 +3996,7 @@ dsr1-fp4-gb200-dynamo-trt:
num-worker: 4
tp: 8
ep: 8
dp-attn: true
dp-attn: false
- conc-list: [ 333 ]
prefill:
num-worker: 2
Expand Down
13 changes: 13 additions & 0 deletions perf-changelog.yaml
Original file line number Diff line number Diff line change
@@ -1,3 +1,16 @@
- config-keys:
- dsr1-fp8-b200-dynamo-trt
- dsr1-fp8-h200-dynamo-trt
- dsr1-fp4-gb200-dynamo-trt
description:
- "Fix metadata inconsistencies in nvidia-master.yaml - TP/EP/DP-attn values now match actual recipe files"
- "B200 FP8 TRT 8K/1K: prefill_ep 8→1 (15 entries), prefill_dp_attn true→false (1 entry)"
- "H200 FP8 TRT 1K/1K: prefill_dp_attn false→true (9 entries)"
- "H200 FP8 TRT 8K/1K: prefill_dp_attn true→false (8 entries)"
- "GB200 FP4 TRT 8K/1K: decode_dp_attn true→false (2 entries)"
- "All fixes are metadata-only; no recipe files were modified"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/919

- config-keys:
- kimik2.5-int4-mi325x-vllm
description:
Expand Down
Loading