Skip to content

[Klaud Cold] kimik2.6-fp4-b200-dynamo-vllm: remove no-enable-flashinfer-autotune, target b200-new / 移除 no-enable-flashinfer-autotune #2445

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
xinli-sw wants to merge 3 commits into from
Closed

[Klaud Cold] kimik2.6-fp4-b200-dynamo-vllm: remove no-enable-flashinfer-autotune, target b200-new / 移除 no-enable-flashinfer-autotune #2445

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
32a9df2
kimik2.6 b200 disagg: remove no-enable-flashinfer-autotune, runner b2…
xinli-sw Aug 1, 2026
92e4f91
fill pr-link
xinli-sw Aug 1, 2026
67241fd
revert runner to b200-multinode
xinli-sw Aug 1, 2026
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
2 changes: 0 additions & 2 deletions ...node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-1p1d-dep4-dep8-c1024.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -95,7 +94,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
2 changes: 0 additions & 2 deletions ...node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-1p1d-dep8-dep8-c2048.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -95,7 +94,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
2 changes: 0 additions & 2 deletions ...lti_node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-1p1d-dep8-tp8-c1.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -93,7 +92,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
2 changes: 0 additions & 2 deletions ...i_node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-1p4d-dep4-tp4-c512.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -92,7 +91,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
2 changes: 0 additions & 2 deletions ...i_node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-1p8d-dep4-tp4-c128.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -92,7 +91,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
2 changes: 0 additions & 2 deletions ...ti_node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-1p8d-dep4-tp4-c32.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -92,7 +91,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
2 changes: 0 additions & 2 deletions ...node/srt-slurm-recipes/vllm/kimi-k2.6/b200-fp4/8k1k/disagg-b200-2p1d-dep8-dep8-c8192.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,7 +75,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
attention-backend: FLASHINFER_MLA
block-size: 128
attention-config: '{"mla_prefill_backend": "TRTLLM_RAGGED"}'
Expand All @@ -95,7 +94,6 @@ backend:
safetensors-load-strategy: prefetch
trust-remote-code: true
no-enable-prefix-caching: true
no-enable-flashinfer-autotune: true
async-scheduling: true
attention-backend: FLASHINFER_MLA
block-size: 128
Expand Down
6 changes: 6 additions & 0 deletions perf-changelog.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5355,3 +5355,9 @@
- "Apply the accuracy-gated Kimi-K2.5 MXFP4 settings: tuned AITER MXFP4 MoE, fused shared experts, FP8 KV cache, block size 16, 16384 batched tokens, 512 sequences, async scheduling, gpu-memory-utilization 0.85 (headroom for CUDA-graph capture on MI355X), and the AITER BF16 GEMM path"
- "Extend the TP4 and TP8 8k1k concurrency sweep from 64 to 128 (1k1k deprecated per #2263)"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2213

- config-keys:
- kimik2.6-fp4-b200-dynamo-vllm
description:
- "Remove no-enable-flashinfer-autotune from Kimi K2.6 B200 disagg recipes, target runner b200-new"
pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2445
Loading