vllm-gpt-oss-optimization

Community

PR-backed GPT-OSS optimization for vLLM

AuthorBBuf
Version1.0.0
Installs0

System Documentation

What problem does it solve?

GPT-OSS optimization in vLLM is enabled through a PR-driven manual that supports auditing, debugging, extending, and documenting GPT-OSS MoE, MXFP4/FP8 quantization, DP/EP, reasoning parser, tool calling, and Eagle spec decode.

Core Features & Use Cases

  • PR-diff driven optimization workflow for GPT-OSS in vLLM, enabling rapid tracking of changes across PRs.
  • Centralized evidence and history linkage for landed PRs, diff dossiers, and model-pr-diff-dossier references.
  • Validation guidance and runtime surfaces to ensure changes in MoE, quantization, and tool parsing behave correctly.

Quick Start

Audit a GPT-OSS PR in vLLM by following the diff-dossier workflow to document changes and validate their impact on MoE, quantization, and tooling.

Dependency Matrix

Required Modules

None required

Components

references

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: vllm-gpt-oss-optimization
Download link: https://github.com/BBuf/AI-Infra-Auto-Driven-SKILLS/archive/main.zip#vllm-gpt-oss-optimization

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.