Vllm 0.20#49
Merged
Merged
Conversation
✅ AccelMark Validation: All submissions validSee the workflow run for details. |
Adds the AccelMark runner for the upcoming vLLM 0.20.x major release on NVIDIA GPUs. This is the supersessor of nvidia_vllm_47f5d58e (vLLM 0.7.3) and tracks the 2026 stack: torch 2.11, CUDA 13.0 (12.8 still supported), HuggingFace Transformers v5, FlashAttention 4 MLA prefill, Model Runner V2, and TurboQuant 2-bit KV cache. What is included: * runners/nvidia_vllm_0c1710bd/ — runner.py, meta.json (with supersedes_chain pointing at the 0.7.3 runner and suite_support self-declaration), requirements.txt, README.md * configs/runner_configs/runner_nvidia_vllm_0c1710bd.yaml.example The README platforms matrix updates automatically from meta.json — no shared file is touched. The 0.7.3 runner remains in the matrix until a successful smoke result on 0.20 lands, at which point its meta.json will gain deprecated_by in a follow-up PR. Capability flags: * SUPPORTED_QUANTIZATION_BACKENDS adds 'turboquant' on top of fp8 / compressed-tensors / gptq_marlin. * Framework version string now reports vllm + transformers together so v4 vs v5 transformers transitions are visible in result.json. * Reuses EngineArgs-field filtering to absorb new 0.20 engine kwargs without breaking old configs. Initial commit, not yet validated on hardware; all suites are marked "pending" in suite_support. Co-authored-by: Cursor <cursoragent@cursor.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Type of change
Testing
# Commands used to verifyChecklist
result.jsonfiles (or I have explained the migration path)BenchmarkRunner, produces validresult.json, includes a reference resultvalidate_submission.pyupdated and all existing results still validateleaderboard/generate.pyproduces correct output on existing resultsRelated issues