{"author":"GodelNumbering","children":[],"created_at":"2026-10-05T15:35:29.000Z","created_at_i":1791214529,"id":49966238,"options":[],"parent_id":null,"points":6,"story_id":49966238,"text":"Purely a hobby side project to see how far I can push a really small model, using (mostly) automated training pipelines<p>Original mention: <a href=\"https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=49869735\">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=49869735</a><p>There were a bunch of requests to release it. Full synthetic data: <a href=\"https:&#x2F;&#x2F;huggingface.co&#x2F;datasets&#x2F;dirac-run&#x2F;ec-training-data\" rel=\"nofollow\">https:&#x2F;&#x2F;huggingface.co&#x2F;datasets&#x2F;dirac-run&#x2F;ec-training-data</a><p>Models: <a href=\"https:&#x2F;&#x2F;huggingface.co&#x2F;dirac-run&#x2F;ec-1.5b-gguf\" rel=\"nofollow\">https:&#x2F;&#x2F;huggingface.co&#x2F;dirac-run&#x2F;ec-1.5b-gguf</a> and <a href=\"https:&#x2F;&#x2F;huggingface.co&#x2F;dirac-run&#x2F;ec-0.6b-gguf\" rel=\"nofollow\">https:&#x2F;&#x2F;huggingface.co&#x2F;dirac-run&#x2F;ec-0.6b-gguf</a><p>Cli <a href=\"https:&#x2F;&#x2F;github.com&#x2F;dirac-run&#x2F;ec\" rel=\"nofollow\">https:&#x2F;&#x2F;github.com&#x2F;dirac-run&#x2F;ec</a><p>feel free to train&#x2F;use the data as you wish.","title":"Show HN: I finetuned 1.5B Qwen to near GPT-4o level bash generation perf","type":"story","url":"https://dirac.run/posts/easycommand"}
