---
title: MinCoder-4B-Expert
canonical_url: "https://www.modelscope.cn/models/beyoru-ai/MinCoder-4B-Expert"
md_url: "https://www.modelscope.cn/models/beyoru-ai/MinCoder-4B-Expert.md"
repository: beyoru-ai/MinCoder-4B-Expert
last_updated: 2026-05-08
license: apache-2.0
model_type:
  - qwen3
architectures:
  - Qwen3ForCausalLM
base_model:
  - beyoru/EvolLLM
base_model_relation: finetune
parameters: 4.0B
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 15
stars: 0
tags:
  - text-generation-inference
  - transformers
  - qwen3
  - code
  - tool
  - agent
  - evolution
  - merge
  - RL
  - grpo
---

# MinCoder-4B-Expert

> MinCoder-4B-Expert - beyoru-ai 在 ModelScope 开源的模型。This model is fine-tuned Qwen model using a custom reinforcement learning (RL) framework that rewards the model for producing solutions passing automated test cases — similar to the process of programming…

beyoru-ai/MinCoder-4B-Expert 是 ModelScope 魔搭社区上的 4.0B 参数机器学习模型，采用 apache-2.0 许可，基于 beyoru/EvolLLM 构建。

- **Repository**: beyoru-ai/MinCoder-4B-Expert
- **License**: apache-2.0
- **Parameters**: 4.0B
- **Base model**: beyoru/EvolLLM
- **Tags**: text-generation-inference, transformers, qwen3, code, tool, agent, evolution, merge, RL, grpo
- **Downloads**: 15
- **Stars**: 0
- **Last updated**: 2026-05-08

Source: https://www.modelscope.cn/models/beyoru-ai/MinCoder-4B-Expert

---

This model is fine-tuned Qwen model using a custom reinforcement learning (RL) framework that rewards the model for producing solutions passing automated test cases — similar to the process of programming task evaluation on LeetCode.

<p align="center">
  <img src="https://cdn-uploads.huggingface.co/production/uploads/65905af887944e494e37e09a/s4drmYGEYWZyt2ZUkxIpI.png" width="300">
</p>


Instead of relying on labeled ground truth answers, the model learns through test-case-based rewards, promoting generalization and reasoning ability in algorithmic problem-solving.