---
license: apache-2.0
language:
- en
- zh
base_model:
- Qwen/Qwen3-8B
tags:
- deep-research
- qwen3
- sft
- rl
- agent
- web-search
library_name: transformers
pipeline_tag: text-generation
---

# QREF — Deep-Research Agent (Qwen3-8B SFT + RL)

Final checkpoints of our deep-research / web-agent pipeline, built on top of **Qwen3-8B**.

| Subfolder | Stage | Training step |
|---|---|---|
| [`SFT/`](./tree/main/SFT) | Supervised Fine-Tuning | `global_step = 1099` |
| [`RL/`](./tree/main/RL)   | Reinforcement Learning | `global_step = 166`  |

Both checkpoints share the same architecture (`Qwen3ForCausalLM`, bfloat16) and tokenizer as the base model.

## License

Apache-2.0, following the base model's license.