RepoNVIDIANVIDIApublished Jun 6, 2026seen 2d

NVIDIA/srt-slurm-recipes

Python

Open original ↗

Captured source

source ↗
published Jun 6, 2026seen 2dcaptured 16hhttp 200method plain

NVIDIA/srt-slurm-recipes

Description: Official NVIDIA/srt-slurm sweep configs for benchmarking LLMs across NVIDIA GPUs and frameworks, spanning aggregated and disaggregated serving on single- and multi-node setups.

Language: Python

License: Apache-2.0

Stars: 0

Forks: 0

Open issues: 0

Created: 2026-06-06T16:27:18Z

Pushed: 2026-06-08T21:38:28Z

Default branch: main

Fork: no

Archived: no

README:

srt-slurm-recipes

This repository is the source of truth for official srt-slurm sweep configurations across node counts, models, GPU platforms, and inference frameworks.

The upstream tool is NVIDIA/srt-slurm. Keep benchmark execution logic in that project; keep reusable sweep configuration in this repository.

Layout

Recipes are organized by benchmark shape:

recipes/
single-node/
/
/
/
multi-node/
/
/
/

Each leaf directory is reserved for one or more *.yaml sweep configs for that exact combination.

Current recipes:

Run python scripts/update_recipe_table.py after adding or removing recipe directories. The same check is available as a pre-commit hook at .githooks/pre-commit. Enable it for a checkout with scripts/install_hooks.sh.

Legend:

  • ✅ Recipe is present
  • ⚠️ Recipe is in progress
  • ❌ Recipe is not working/not planned

Multi-Node

Model GB200 / GB300

MiniMax-M2.5 ✅

Single-Node

Model B200 / B300 H100

DeepSeek-R1 ✅ ⚠️

DeepSeek-V4-Pro ✅ ⚠️

gpt-oss-120b ✅ ✅

Kimi-K2.5 ✅ ⚠️

MiniMax-M2.5 ✅ ⚠️

NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 ✅ ⚠️

Qwen3.5-397B-A17B-FP8 ✅ ⚠️

Running a Sweep

Install srt-slurm from upstream, then point srtctl at a recipe YAML in this repo:

git clone https://github.com/NVIDIA/srt-slurm.git
cd srt-slurm
uv pip install -e .

RECIPES_PATH=/path/to/srt-slurm-recipes
FRAMEWORK=
uv run srtctl dry-run -f "${RECIPES_PATH}/recipes/single-node/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4/B200/${FRAMEWORK}/1k1k.yaml"
uv run srtctl apply -f "${RECIPES_PATH}/recipes/single-node/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4/B200/${FRAMEWORK}/1k1k.yaml" \
--tags official,NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4,B200,"${FRAMEWORK}"

The dry-run command should be used before submitting jobs.

Contributing

See [CONTRIBUTING.md](.github/CONTRIBUTING.md) for how to add or update official sweep recipes.

Support, Security & Governance

  • Support level: Maintained — see [SUPPORT.md](.github/SUPPORT.md) for how to get help.
  • Security: Report vulnerabilities privately per [SECURITY.md](SECURITY.md). Do not file public issues for security reports.
  • Code of Conduct: [CODE_OF_CONDUCT.md](.github/CODE_OF_CONDUCT.md)
  • Governance & Maintainers: [GOVERNANCE.md](.github/GOVERNANCE.md) · [MAINTAINERS.md](.github/MAINTAINERS.md)
  • Citation: [CITATION.md](.github/CITATION.md)

License

This repository follows the upstream srt-slurm license. See [LICENSE](LICENSE).

Notability

Scored, but no written rationale attached yet.

NVIDIA has a repo signal matching evals and quality, infrastructure.