vllm-project/llm-compressor 3,931
vllm-project/guidellm 1,662
GuideLLM resolves torch from the CPU index and its own comparison table omits its vLLM backend
vava-nessa/free-coding-models 2,798
Free coding models, measured by p95 and jitter, at a cost paid in your provider quota
ulab-uiuc/LLMRouter 3,045
LLMRouter's extra named all contains exactly the GPU extra
Trystan-SA/claude-design-system-prompt 2,027
A design system prompt that documents how to weaken itself for older models
TheBlewish/Automated-AI-Web-Researcher-Ollama 3,022
The first thing in this researcher's readme is a banner for a different repository
Sumanth077/ai-engineering-toolkit 3,393
ai-engineering-toolkit is two files, and its best feature is a licence column
StartupHakk/OpenMonoAgent.ai 1,904
OpenMonoAgent: a 12-step pipeline with ten steps written down
stanford-crfm/helm 2,958
HELM entered maintenance mode in June 2026, and its package name is not its binary name
spider-rs/spider 2,812
Spider starts with HTTP and only opens Chrome when a page needs it
sooryathejas/METATRON 4,235
METATRON: a two-terminal pen-testing assistant with the database password in the README
skyllwt/AutoSci 1,692
AutoSci keeps the paper on a frozen branch and ships four runtime branches instead
signerless/llm-checker 3,046
llm-checker: AI-powered in the subtitle, deterministic in the scoring core
sgl-project/mini-sglang 5,479
Mini-SGLang: reading a 5,000 line LLM server, and its platform ceiling
sdyckjq-lab/llm-wiki-skill 2,574
llm-wiki compiles knowledge once instead of re-deriving it
relation-graph/relation-graph 2,281
relation-graph: five packages, one from a CDN, and a duplicated examples list
rednote-machine-learning/RedKnot 3,029
RedKnot: the headline saving excludes memory traffic, kernel launches and interconnect
QwenLM/Qwen3-Omni 4,075