Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding

<p><strong><a href="https://deep-reinforce.com/ornith_1_0.html">Ornith-1.0: Self-Scaffolding LLMs for Agentic Coding</a></strong></p> This is an interesting new open weights (MIT licensed) model, the first model release from DeepReinforce.</p> <blockquote> <p>[...] with variants including 9B Dense, 31B Dense, 35B MoE, and 397B MoE. Built on top of pretrained Gemma 4 and Qwen 3.5, it achieves state-of-the-art performance among open-source models of comparable size on coding benchmarks.</p> </blockquote> <p>As far as I can tell the licenses of those underlying models is compatible with being used in this way - Gemma 4 is Apache 2.0 licensed (and not bound by the janky additional <a href="https://ai.google.dev/gemma/terms">Gemma Terms of Use</a> that afflicted the previous Gemma models) and Qwen 3.5 is Apache 2.0 licensed as well.</p> <p>I've been running the model using LM Studio and the <a href="https://huggingface.co/deepreinforce-ai/Ornith-1.0-35B-GGUF">ornith-1.0-35b-Q4_K_M.gguf</a> (20GB) GGUF, hooked up to <a href="https://pi.dev/">Pi</a>. Initial impressions are very good - it seems to be able to run the agent harness over many tool calls in a proficient way.</p> <p>Here's <a href="https://gisthost.github.io/?35da4d9ce7f0c27124c67655a0dc9e5d">a terminal session</a> where I asked it to "find the code that decodes the actor cookie" and then "find the code that opens the insert dialog when thebutton is clicked" against a Datasette checkout, which it handled with ease.</p> <p>I also had it <a href="https://gist.github.com/simonw/1869e1bbcafe5bcad0f26351f6a978a6">draw this pelican</a>, which came out at 103 tokens/second:</p> <p><img alt="Cartoon illustration of a white pelican (albeit slightly mangled) with a large orange beak riding a red bicycle across green hills. The scene has a blue sky with a yellow sun and three white clouds, and small grass tufts dot the foreground." src="https://static.simonwillison.net/static/2024/ornith-1-pelican.png" /></p> <p>It's a li