SALEETAI/Coding-Agent

1---2title: Coding Agent3emoji: 🏃4colorFrom: yellow5colorTo: green6sdk: docker7pinned: false8license: apache-2.09---10 11 12---13title: Coding Agent Qwen-SFT v314emoji: 🚀15colorFrom: blue16colorTo: gray17sdk: docker18pinned: false19license: apache-2.020---21 22# 🚀 Coding Agent Qwen-SFT v2 (GGUF)23 24This is a fine-tuned **Qwen2.5-Coder-7B** model, optimized for production-grade Python, C++, and Rust development. 25 26### 🌟 Key Features27- **Logic-First:** Trained to avoid "Textbook" descriptions and provide direct code solutions.28- **Efficient:** Quantized to 4-bit (GGUF) to run smoothly on standard CPUs.29- **Context-Aware:** Handles complex concurrency, memory safety, and architectural prompts.30 31### 💻 Use it Locally (Ollama)32You can run this exact model on your own machine:331. **Install [Ollama](https://ollama.com)**342. **Download the GGUF file** from this repository.353. **Create a `Modelfile`**:36   ```dockerfile37   FROM ./coding-agent-qwen-sft-v2-GGUF.q4_k_m.gguf38   PARAMETER temperature 0.439   PARAMETER repeat_penalty 1.240   ```414. **Run:** `ollama create coding-agent -f Modelfile` then `ollama run coding-agent`42 43### 🛠️ Inference Specs44- **Repetition Penalty:** 1.245- **Temperature:** 0.446- **Stop Token:** `<|im_end|>`47 48 49Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference50