Ask HN: Open-Source Coding Model and Harness at Claude Sonnet / Opus Level Perf?
hackernews
|
|
{'이벤트': '📰', '머신러닝/연구': '📰', '하드웨어/반도체': '📰', '취약점/보안': '📰', '기타 AI': '📰', 'AI 딜': '📰', 'AI 모델': '📰', 'AI 서비스': '📰', 'discount': '📰', 'news': '📰', 'review': '📰', 'tip': '📰'} AI 서비스
#ai 서비스
#review
#구글 플레이 스토어
#리뷰
#앱
#앱 리뷰
#키워드 검색
요약
구글 플레이 스토어에 사용자들이 앱 리뷰를 쉽게 탐색할 수 있는 '키워드 검색' 기능이 새롭게 도입되었습니다. 이를 통해 수많은 리뷰 속에서 특정 단어를 입력해 자신이 원하는 기능이나 문제점과 관련된 평가를 빠르게 찾아볼 수 있게 되었습니다. 결과적으로 사용자들은 개별 앱의 장단점을 보다 정확하게 파악하여 앱 설치 시 합리적인 결정을 내리는 데 큰 도움을 받을 수 있습니다.
왜 중요한가
본문
I am tired of Anthropic's rate limits. Is there a coding model + coding harness combination that I can run 100% locally? I need a local AI model and coding harness that matches Claude Code in performance (ideally Opus-level, but Sonnet-level would also be amazing). With Claude Code's Mar 31 leak, we now have free, open source competing harnesses like https://github.com/ultraworkers/claw-code and https://github.com/Gitlawb/openclaude
Has anyone found a 100% local setup that rivals Claude's models in quality for professional-level agentic coding?<p>Some Context:
- I have about 20GB-30GB of space left on my Macbook Pro, which I can use for downloading a local model. I do not want to use Ollama Cloud and pay yet another $20/mo subscription.
- I have tried Gemma 4 on Ollama (Google's SOTA open source model that released Apr 2026), and it is absolute garbage. Stay away.</p>
Has anyone found a 100% local setup that rivals Claude's models in quality for professional-level agentic coding?<p>Some Context:
- I have about 20GB-30GB of space left on my Macbook Pro, which I can use for downloading a local model. I do not want to use Ollama Cloud and pay yet another $20/mo subscription.
- I have tried Gemma 4 on Ollama (Google's SOTA open source model that released Apr 2026), and it is absolute garbage. Stay away.</p>