Strongest CodeScout model — open-source SOTA on SWE-Bench code localization. CodeScout-14B is part of the CodeScout family of open-source RL-trained code search agents. CodeScout models achieve state-of-the-art repository-level code localization using nothing more than a standard Unix terminal — no static analysis, no repository graphs, no language-specific tooling. — Open-source SOTA on SWE-Bench Verified, Pro, and Lite for code localization — Outperforms 2–18× larger base and post-trained LLMs across all benchmarks — Surpasses GPT-5 and approaches Claude Sonnet 4.5 using RepoNavigator, despite using only a bash terminal — Achieves 8–33% higher function-level F1 than Qwen3-32B (Thinking) Performance on SWE-Bench code localization (instance-averaged F1 scores): Code localization performance on SWE-Bench Verified. CodeScout (⭐) achieves superior or competitive results over larger open-source LLMs and narrows the gap with closed-source frontier models. CodeScout-14B is trained directly from Qwen3-14B using GSPO (Group Sequence Policy Optimization) reinforcement learning. — Training data: 9,600 instances from SWE-Smith (39K filtered, 128 repos) — RL steps: 300 — Batch size: 32, with…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: OpenHands
Теги: qwen3, code-search, code-localization, reinforcement-learning, agent, software-engineering, GSPO, OpenHands
Лайков: 4 | Загрузок: 156
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
