docs: honest base_model attribution, drop marketing

README.md

@@ -1,5 +1,6 @@
 ---
 license: apache-2.0
+base_model: Qwen/Qwen3-Reranker-8B
 library_name: transformers
 pipeline_tag: text-classification
 language:
@@ -11,25 +12,21 @@ tags:
 - zenlm
 - hanzo
 - reranker
+- retrieval
 ---
 
 # Zen Reranker 8B
 
-8B-parameter cross-encoder reranker for retrieval pipelines and search relevance. Part of the Zen reranker family.
+8B-parameter cross-encoder reranker for retrieval pipelines and search relevance — re-scores candidate passages for RAG and search quality. Part of the Zen reranker line.
 
-## Hosted via Hanzo gateway
+Repackaged from [Qwen/Qwen3-Reranker-8B](https://huggingface.co/Qwen/Qwen3-Reranker-8B) (apache-2.0, Alibaba Qwen). Native HuggingFace `safetensors`, re-hosted under the Zen reranker line. **Not trained from scratch** — a permissively-licensed redistribution for the OSS-clean Zen model line.
 
-Served at `api.hanzo.ai` as `zen-reranker-8b`.
+## Usage
 
-## Files
+Score query/passage pairs following the upstream [Qwen/Qwen3-Reranker-8B](https://huggingface.co/Qwen/Qwen3-Reranker-8B) recipe (yes/no token scoring).
 
-Native HuggingFace `safetensors` weights, loadable directly with `transformers`:
+Hosted via the Hanzo gateway (`api.hanzo.ai`) as `zen-reranker-8b`. GGUF build for CPU inference: [zenlm/zen-reranker-8B-GGUF](https://huggingface.co/zenlm/zen-reranker-8B-GGUF).
 
-```python
-from transformers import AutoModel, AutoTokenizer
-m = AutoModel.from_pretrained("zenlm/zen-reranker-8B")
-t = AutoTokenizer.from_pretrained("zenlm/zen-reranker-8B")
-```
-
-For GGUF / Ollama deployment, see [zenlm/zen-reranker-8B-GGUF](https://huggingface.co/zenlm/zen-reranker-8B-GGUF).
+## License
 
+`apache-2.0`. Upstream: **Qwen/Qwen3-Reranker-8B** by Alibaba Qwen. This repository redistributes the weights under the same license.