by ywchiu · Agent Tool · ★ 26
Agentic Coding Benchmark 中文版 (Traditional Chinese) An automated benchmark suite for evaluating LLM agentic coding ability via OpenRouter tool-use API. Give a model a vague prompt and 4 tools (writefile, readfile, runcommand, listfiles), see if it builds something that actually works. Claude models are tested via Claude Code (native API); Gemma 4 is tested via Google's Gemini native API. Why these models? This benchmark is designed to find the best bang-for-your-buck models for agentic coding.
| Stars | 26 |
| Forks | 4 |
| Language | Shell |
| Category | Agent Tool |
| Quality Score | 51.6013484049432/100 |
| Last Updated | 2026-04-23 |
| Created | 2026-03-19 |
| Platforms | cli |
| Est. Tokens | ~145k |
These tools work well together with local_agentic_llm for enhanced workflows:
Explore other popular agent tool tools:
local_agentic_llm is local_agentic_llm. It is categorized as a Agent Tool with 26 GitHub stars.
local_agentic_llm is primarily written in Shell.
You can find installation instructions and usage details in the local_agentic_llm GitHub repository at github.com/ywchiu/local_agentic_llm. The project has 26 stars and 4 forks, indicating an active community.