TAAFT
Free mode
100% free
Freemium
Free Trial
Deals

SERA 32B GA

By Ai2
Model family: SERA
SERA-32B-GA is the second Open Coding Agents model from Ai2, using the same Qwen 3-32B base as SERA-32B but with GLM-4.5-Air as the teacher in the Soft Verified Generation two-rollout pipeline. It is trained on 16k synthetic coding-agent trajectories and reaches around 46.6% on SWE-bench Verified at 32K context. The model provides a high-performing, Apache-2.0 licensed coding agent whose data and training recipe are available for further research and specialization on private repositories.
Text Gen 7
Released: January 27, 2026

Overview

SERA-32B-GA is a 32B Qwen 3 based coding agent variant trained on GLM-4.5-Air trajectories, delivering about 46.6% SWE-bench Verified performance as a slightly smaller-teacher alternative to SERA-32B.

About Ai2

Ai2 is a 501(c)(3) non-profit AI research institute founded in 2014 by the late Paul Allen (Microsoft co-founder), dedicated to conducting high-impact, open AI research and engineering for the common good, including open language models (OLMo), scientific AI tools (Semantic Scholar, Asta), environmental AI platforms, and embodied robotics research.

Industry: Artificial Intelligence
Company Size: 313
Location: Seattle, Washington, US
Website: allenai.org
View Company Profile

Tools using SERA 32B GA

No tools found for this model yet.

Last updated: April 6, 2026
0 AIs selected
Clear selection
#
Name
Task