Lynn: open-source desktop/CLI AI Agent with GUI Session Map, headless workers, realtime voice, long-term memory, Brain V2 routing, local 27B/35B GGUF support, and contract-based Agent Regression Kit.

Benchmark of local LLMs (Qwen 3.6, Gemma 4) vs ChatGPT and Gemini on coding tasks. Apple M5, 32GB. Methodology, raw outputs, judge prompts, scores, and charts.

Saotri Bench — coding benchmark for evaluating LLM agents on multi-phase programming tasks with hidden requirements.

Pokedex benchmark for LLMs

面向 AI IDE / 编码 Agent 的编程能力基准测试框架，内置防背题三防线：公私用例分离、泛化差距检测、参数化生成

Raw logs of Claude Code running on local Qwen3.5-27B (llama.cpp). Builds a Python todo app with 50 tests. Real-world performance data: 30 min, cache thrashing, 38 t/s generation.