Discussion

• @soldni Luca Soldaini on x

  scaling laws for agent specialization are so cool-unsurprisingly, you can get much better multipliers if you do repo specific trajectory generation great work @Tim_Dettmers @ethnlshn! [image]

• @ethnlshn Ethan Shen on x

  Today, we release SERA-32B, an approach to coding agents that matches Devstral 2 at just $9,000. It is fully open-source and you can train your own model easily - at 26x the efficiency of using RL. Paper: https://allenai.org/... Here's how 🧵

• @ethnlshn Ethan Shen on x

  SERA is based on two key innovations: 1. You don't need unit tests to generate high-quality code data. 2. Unverified coding data can be equal to verified data. This enables SoTA performance and new repository agents that outperform frontier open models on personal codebases. [ima…

• @allen_ai @allen_ai on x

  We're releasing: ✅ A family of strong open coding models ✅ SERA, our training method for building your own agents ✅ Code, recipes, data, + Claude Code integration Here's how to get started with SERA via Claude Code: https://www.youtube.com/...

• @tim_dettmers Tim Dettmers on x

  We release SERA, the first model part of Ai2's Open Coding Agent series. SERA is a SoTA agent for its size, super simple, and 26x more efficient than RL. In my blog post, I write about my personal journey of building this coding agent: https://timdettmers.com/... Details: 👇

• @nvidiaaidev @nvidiaaidev on x

  🎉 Congrats to @allen_ai on your just released Open Coding Agents. 🏎️ Turbocharged on NVIDIA GPUs, their fully open models are achieving state‑of‑the‑art SWE‑Bench Verified performance for open systems. What sets this model apart is that it is fully open source and easily

• @allen_ai @allen_ai on x

  Introducing Ai2 Open Coding Agents—starting with SERA, our first-ever coding models. Fast, accessible agents (8B-32B) that adapt to any repo, including private codebases. Train a powerful specialized agent for as little as ~$400, & it works with Claude Code out of the box. 🧵 [ima…