-
Notifications
You must be signed in to change notification settings - Fork 208
[NV] Add MiniMax-M3 FP8 B300 Dynamo vLLM recipes #1788
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.
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
Closed
Changes from all commits
Commits
Show all changes
2 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
80 changes: 80 additions & 0 deletions
80
...marks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b300-fp8/1k1k/1p1d-dep2-tep8-1k1k.yaml
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,80 @@ | ||
| name: "minimax-m3-vllm-disagg-b300-1p1d-fp8-dep2-tep8-1k1k" | ||
|
|
||
| model: | ||
| path: "MiniMaxAI/MiniMax-M3-MXFP8" | ||
| container: "vllm/vllm-openai:minimax-m3" | ||
| precision: "fp8" | ||
|
|
||
| resources: | ||
| gpu_type: "b300" | ||
| gpus_per_node: 8 | ||
| prefill_nodes: 1 | ||
| decode_nodes: 1 | ||
| prefill_workers: 1 | ||
| decode_workers: 1 | ||
| gpus_per_prefill: 2 | ||
| gpus_per_decode: 8 | ||
|
|
||
| dynamo: | ||
| install: true | ||
| version: 1.3.0.dev20260614 | ||
|
|
||
| frontend: | ||
| type: dynamo | ||
| enable_multiple_frontends: false | ||
|
|
||
| backend: | ||
| type: vllm | ||
| connector: null | ||
|
|
||
| prefill_environment: | ||
| UCX_NET_DEVICES: "all" | ||
| UCX_TLS: "rc,cuda_ipc,cuda_copy,sm,self,tcp" | ||
| VLLM_FLOAT32_MATMUL_PRECISION: "high" | ||
|
|
||
| decode_environment: | ||
| UCX_NET_DEVICES: "all" | ||
| UCX_TLS: "rc,cuda_ipc,cuda_copy,sm,self,tcp" | ||
| VLLM_FLOAT32_MATMUL_PRECISION: "high" | ||
|
|
||
| vllm_config: | ||
| prefill: | ||
| tensor-parallel-size: 1 | ||
| data-parallel-size: 2 | ||
| data-parallel-rpc-port: 13345 | ||
| trust-remote-code: true | ||
| no-enable-prefix-caching: true | ||
| kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}' | ||
| block-size: 128 | ||
| gpu-memory-utilization: 0.90 | ||
| max-model-len: 2304 | ||
| language-model-only: true | ||
| stream-interval: 32 | ||
| max-cudagraph-capture-size: 2048 | ||
| max-num-batched-tokens: 2048 | ||
|
|
||
| decode: | ||
| tensor-parallel-size: 8 | ||
| enable-expert-parallel: true | ||
| trust-remote-code: true | ||
| no-enable-prefix-caching: true | ||
| kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}' | ||
| block-size: 128 | ||
| gpu-memory-utilization: 0.90 | ||
| max-model-len: 2304 | ||
| language-model-only: true | ||
| stream-interval: 32 | ||
| max-num-seqs: 4096 | ||
| max-num-batched-tokens: 16384 | ||
| max-cudagraph-capture-size: 8196 | ||
|
|
||
| health_check: | ||
| max_attempts: 360 | ||
| interval_seconds: 10 | ||
|
|
||
| benchmark: | ||
| type: "sa-bench" | ||
| isl: 1024 | ||
| osl: 1024 | ||
| concurrencies: "4x16x64x128x4096" | ||
| req_rate: "inf" | ||
81 changes: 81 additions & 0 deletions
81
...marks/multi_node/srt-slurm-recipes/vllm/minimax-m3/b300-fp8/1k1k/1p2d-dep2-dep4-1k1k.yaml
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,81 @@ | ||
| name: "minimax-m3-vllm-disagg-b300-1p2d-fp8-dep2-dep4-1k1k" | ||
|
|
||
| model: | ||
| path: "MiniMaxAI/MiniMax-M3-MXFP8" | ||
| container: "vllm/vllm-openai:minimax-m3" | ||
| precision: "fp8" | ||
|
|
||
| resources: | ||
| gpu_type: "b300" | ||
| gpus_per_node: 8 | ||
| prefill_nodes: 1 | ||
| decode_nodes: 1 | ||
| prefill_workers: 1 | ||
| decode_workers: 2 | ||
| gpus_per_prefill: 2 | ||
| gpus_per_decode: 4 | ||
|
|
||
| dynamo: | ||
| install: true | ||
| version: 1.3.0.dev20260614 | ||
|
|
||
| frontend: | ||
| type: dynamo | ||
| enable_multiple_frontends: false | ||
|
|
||
| backend: | ||
| type: vllm | ||
| connector: null | ||
|
|
||
| prefill_environment: | ||
| UCX_NET_DEVICES: "all" | ||
| UCX_TLS: "rc,cuda_ipc,cuda_copy,sm,self,tcp" | ||
| VLLM_FLOAT32_MATMUL_PRECISION: "high" | ||
|
|
||
| decode_environment: | ||
| UCX_NET_DEVICES: "all" | ||
| UCX_TLS: "rc,cuda_ipc,cuda_copy,sm,self,tcp" | ||
| VLLM_FLOAT32_MATMUL_PRECISION: "high" | ||
|
|
||
| vllm_config: | ||
| prefill: | ||
| tensor-parallel-size: 1 | ||
| data-parallel-size: 2 | ||
| data-parallel-rpc-port: 13345 | ||
| trust-remote-code: true | ||
| no-enable-prefix-caching: true | ||
| kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}' | ||
| block-size: 128 | ||
| gpu-memory-utilization: 0.90 | ||
| max-model-len: 2304 | ||
| language-model-only: true | ||
| stream-interval: 32 | ||
| max-cudagraph-capture-size: 2048 | ||
| max-num-batched-tokens: 2048 | ||
|
|
||
| decode: | ||
| tensor-parallel-size: 1 | ||
| data-parallel-size: 4 | ||
| data-parallel-rpc-port: 13345 | ||
| trust-remote-code: true | ||
| no-enable-prefix-caching: true | ||
| kv-transfer-config: '{"kv_connector": "NixlConnector", "kv_role": "kv_both"}' | ||
| block-size: 128 | ||
| gpu-memory-utilization: 0.90 | ||
| max-model-len: 2304 | ||
| language-model-only: true | ||
| stream-interval: 32 | ||
| max-num-seqs: 1024 | ||
| max-num-batched-tokens: 16384 | ||
| max-cudagraph-capture-size: 8196 | ||
|
|
||
| health_check: | ||
| max_attempts: 360 | ||
| interval_seconds: 10 | ||
|
|
||
| benchmark: | ||
| type: "sa-bench" | ||
| isl: 1024 | ||
| osl: 1024 | ||
| concurrencies: "2048" | ||
| req_rate: "inf" |
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🟡 WARNING:
max-cudagraph-capture-size: 8196is likely a typo for8192(2^13)Why it matters: 8196 is not a power of 2. Every other recipe in this repo uses power-of-2 values (e.g.
2048). vLLM uses this as a batch-size threshold for CUDA graph capture — a non-power-of-2 value is unusual and likely unintentional. This same value appears across all 6 decode configs in this PR.Fix: Since these are imported verbatim from NVIDIA/srt-slurm#223, you may want to confirm with upstream whether
8196is intentional or should be8192. If it's an upstream typo, it's worth fixing here too.