README.md

---
title: Alkaid A
emoji: ⭐
colorFrom: indigo
colorTo: purple
sdk: docker
app_port: 7860
pinned: true
license: apache-2.0
---

# Alkaid A — Train & Chat

Fine-tuned from [Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled](https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled).

## Features
- **Train Tab**: Fine-tune the model with LoRA directly in this Space
- **Chat Tab**: Test inference with your trained model
- **Custom Data**: Paste your own JSONL training data

## Requirements
- Hugging Face Pro account (for GPU Spaces)
- A10G or A100 GPU hardware selection
- Write-access HF token for pushing trained models