SERA-14B is built for more setups and easier deployment—a smaller, more accessible option that still keeps SERA's cheap, customizable approach.
We've also revamped the open SERA training data into a general, model-agnostic format that's easier to reuse across different workflows.
What's new:

Verification thresholds per sample

More metadata for filtering & analysis
You can drop in SERA-14B or retrain with our refreshed data. We look forward to seeing what you build!

Model & data:
https://huggingface.co/collections/allenai/open-coding-agents…

Learn more:
https://allenai.org/blog/open-coding-agents…