# SkillGym fine-tunes Qwen3.5-35B-A3B on skill trajectories

A paper proposes SkillGym, which turns human-written skills into 2,756 training environments and fine-tunes on 8,364 successful trajectories. It reports that Qwen3.5-35B-A3B in Claude Code then gains 19.10 points on Terminal-Bench 2.1 and reaches 51.47% on skill-assisted SkillsBench v1.1.

Language: en
Time zone: America/Los_Angeles
Current time (UTC, cached as of): 2026-11-17T15:24:00.000Z
Query date (America/Los_Angeles): 2026-11-17

HTML: https://didcodexreset.com/news/f6635ef4b9914e71a854e51e.html

Content language: en
Localization state: sameLanguage

Source: DAIR.AI · Published 9/27/2026, 15:43:06

A paper proposes SkillGym, which turns human-written skills into 2,756 training environments with code-based checkers and fine-tunes on 8,364 successful trajectories. The paper is [SkillGym: Internalizing Human Skills into LLMs for Real-World Problem Solving](https://academy.dair.ai/papers/skillgym-internalizing-human-skills-into-llms-for-real-world-problem-solving-2609.27717).

After fine-tuning, Qwen3.5-35B-A3B running in Claude Code gains 19.10 points on Terminal-Bench 2.1 and 28.13 points on skill-assisted SkillsBench v1.1, where it reaches 51.47%. The paper reports that result as above the reported scores for Claude Sonnet 4.6 and GPT-5.4 Mini. With no skills loaded, the trained model beats the base model that has the skills in context.

Tags: SkillGym, Qwen, Terminal-Bench, SkillsBench, Fine-tuning

[View original post](https://x.com/dair_ai/status/2104339055194546595)
