Optimized vLLM setup for Qwen3.6-27B-FP8 on dual RTX PRO 6000 Blackwell (192 GB GDDR7, no NVLink) ; config, benchmark sweep results, and custom chat template with thinking mode off by default.
GitHub repository with 10 stars and 0 forks.
Language: Shell
Topics: benchmark, blackwell, fp8, llm-inference, local-llm, multi-token-prediction, qwen3, rtx-pro-6000, speculative-decoding, vllm
2026-06-06: 10 stars and 0 forks.
Open Source Continuous Inference Benchmark Research Platform Kimi K2.6, DeepSeekv4, GLM5 - GB200 NVL72 vs MI355X vs B200 vs GB300 NVL72 & soon™ TPUv6e/v7/Trainium2/3
GitHub repository with 1,099 stars and 195 forks.
Trending score: 1.96; stars gained: +4; forks gained: +1.
Language: Shell
Topics: ai, benchmark, llm, pytorch, sglang, vllm
antirez/ds4 (DwarfStar 4) on NVIDIA DGX Spark — install, benchmarks, and roofline analysis. Steady-state decode at ~95% of bandwidth ceiling; MTP and concurrency analyzed.
GitHub repository with 25 stars and 0 forks.
Trending score: 0.16; stars gained: +0; forks gained: +0.
Language: Shell
Topics: benchmark, cuda, deepseek-v4-flash, dgx-spark, gb10, gguf
Official PHP benchmark suite
GitHub repository with 67 stars and 5 forks.
Trending score: 0.09; stars gained: +0; forks gained: +0.
Language: Shell
Topics: php, benchmark
Real-world infrastructure tasks to evaluate AI agents, starting with Kubernetes.
GitHub repository with 7 stars and 2 forks.
Trending score: 0.08; stars gained: +0; forks gained: +0.
Language: Shell
Topics: agents, ai, benchmark, infrastructure, kubernetes
An agentic skills framework & software development methodology that works.
GitHub repository with 228,369 stars and 20,316 forks.
Trending score: 6.21; stars gained: +1,838; forks gained: +182.
Language: Shell
🐹 Clean, uninstall, analyze, optimize, and monitor your Mac from the terminal.
GitHub repository with 55,994 stars and 1,965 forks.
Trending score: 4.56; stars gained: +190; forks gained: +20.
Language: Shell
Topics: clean, cleaner, cleaner-script, macos, cleaner-cli, mac
A Collection of Awesome Jellyfin Plugins, Themes, Guides and More!
GitHub repository with 8,201 stars and 237 forks.
Trending score: 3.41; stars gained: +110; forks gained: +8.
Language: Shell
Topics: awesome, jellyfin, awesome-list
An open-source Android app to let you browse YouTube and other services freely.
GitHub repository with 5,419 stars and 165 forks.
Trending score: 3.32; stars gained: +89; forks gained: +7.
Language: Shell
Topics: android, mediaplayer, youtube, newpipe, sponsorblock
A Zsh theme
GitHub repository with 54,456 stars and 2,422 forks.
Trending score: 3.30; stars gained: +136; forks gained: -2.
Language: Shell
Node Version Manager - POSIX-compliant bash script to manage multiple active node.js versions
GitHub repository with 93,816 stars and 10,225 forks.
Trending score: 3.28; stars gained: +133; forks gained: -3.
Language: Shell
🔍 The hardest search benchmark in the wild — vague, multi-turn, proactive. 200 long-horizon tasks with persona-driven progressive disclosure, scored by verifiable schema-free knowledge-graph evaluation. No vibes, just triplet F1.
GitHub repository with 1,008 stars and 63 forks.
Trending score: 3.33; stars gained: +50; forks gained: +37.
Language: Python
Topics: agentic-ai, benchmark, llm, proactive-agent, search, search-agent
Internal Safety Collapse (ISC): Turning the LLM or an AI Agent into a sensitive data generator.
GitHub repository with 865 stars and 142 forks.
Trending score: 2.51; stars gained: +11; forks gained: +3.
Language: Python
Topics: agent-safety, ai-safety, benchmark, jailbreak, large-language-models, llm-safety
MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research · 浏览器里运行的安卓模拟器 · Browser-hosted Android Simulator · Verifiable Evaluation · Scalable Online RL Training
GitHub repository with 621 stars and 98 forks.
Trending score: 2.50; stars gained: +12; forks gained: +1.
Language: TypeScript
Topics: benchmark, mobile-agent, reinforcement-learning, vlm, agents, gym
Complete guide to running large language models locally on AMD Strix Halo / Ryzen AI MAX+ 395 with Radeon 8060S (gfx1151) and 96GB/128GB unified memory. Covers BIOS config, Ubuntu/kernel setup, Ollama, llama.cpp Vulkan/RADV, ROCm/HIP, vLLM, and 70B/120B GGUF evidence.
GitHub repository with 143 stars and 6 forks.
Trending score: 1.97; stars gained: +9; forks gained: +0.
Language: Python
Topics: amd, benchmark, gfx1151, llama-cpp, llm, local-llm
Open Source Continuous Inference Benchmark Research Platform Kimi K2.6, DeepSeekv4, GLM5 - GB200 NVL72 vs MI355X vs B200 vs GB300 NVL72 & soon™ TPUv6e/v7/Trainium2/3
GitHub repository with 1,099 stars and 195 forks.
Trending score: 1.96; stars gained: +4; forks gained: +1.
Language: Shell
Topics: ai, benchmark, llm, pytorch, sglang, vllm
OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.
GitHub repository with 7,086 stars and 788 forks.
Trending score: 1.55; stars gained: +3; forks gained: +0.
Language: Python
Topics: evaluation, benchmark, large-language-model, chatgpt, llm, llama2