mini-coder-4b is a 4B parameter model distilled from Qwen 3 Coder 30B A3B. It punches well above its weight, outperforming gpt-oss-120b on SWE-bench Verified Bash only: It is trained on 400k training trajectories using the lightweight mini-swe-agent scaffolding and the SWE-smith dataset of GitHub issues. Unlike existing agentic SWE models, the mini-coder models can be post-trained on a single 80GB GPU—or smaller. They work seamlessly with mini-swe-agent, a lightweight, scalable, and developer-friendly agentic framework well-suited for RL fine-tuning. And because they are dense rather than MoE models, they benefit from a more mature fine-tuning ecosystem. This example shows how to generate SWE-bench trajectories using mini-swe-agent as the agentic scaffolding (recommended) and vLLM as the local inference engine. First, launch a vLLM server with your chosen model. For example: By default, the server will be available at http://localhost:8000. Second, edit the mini-swe-agent SWE-bench config file located in src/minisweagent/config/extra/swebench.yaml to include your local vLLM model: Now you’re ready to generate trajectories! Let’s solve the djangodjango-11099` instance of SWE-bench…
Модальности:
Генерация текста
Области применения:
Генерация кода Диалог / чат
Задача: Генерация текста
Автор: ricdomolm
Теги: qwen3, agent, code, conversational, text-generation-inference, endpoints_compatible
Лайков: 4 | Загрузок: 81
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
