---
tags:
- gguf
- llama.cpp
- unsloth
datasets:
- TeichAI/gpt-5.2-high-reasoning-250x
base_model: TeichAI/Qwen3-8B-GPT-5.2-High-Reasoning-Distill
---

# Qwen3 8B x GPT 5.2 (High Reasoning)

This model was trained on 250 examples from **GPT 5.2** with reasoning effort set to high.

Recent updates fixed a bug in output style, added new line in between reasoning blocks.