zenlm/zen-reranker

Zen Reranker

Zen Reranker is a high-performance reranking model for search and retrieval pipelines. Part of the Zen AI model family by Hanzo AI.

Overview

Zen Reranker is optimized for:

• Retrieval-Augmented Generation (RAG) — re-score retrieved passages for LLM context
• Search quality improvement — rerank initial BM25/dense retrieval results
• Cross-lingual retrieval — strong multilingual performance
• DSO integration — compatible with Hanzo's Decentralized Semantic Optimization

Quick Start

import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification

model_name = "zenlm/zen-reranker"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name, torch_dtype=torch.float16)

def rerank(query, passages):
    pairs = [[query, p] for p in passages]
    inputs = tokenizer(
        pairs, padding=True, truncation=True,
        max_length=512, return_tensors="pt"
    )
    with torch.no_grad():
        scores = model(**inputs).logits.squeeze(-1)
    ranked = sorted(zip(passages, scores.tolist()), key=lambda x: x[1], reverse=True)
    return ranked

query = "What is the capital of France?"
passages = ["Paris is the capital of France.", "Berlin is in Germany.", "Madrid is in Spain."]
results = rerank(query, passages)
for passage, score in results:
    print(f"{score:.3f}: {passage}")

With sentence-transformers

from sentence_transformers import CrossEncoder

model = CrossEncoder("zenlm/zen-reranker")
scores = model.predict([
    ["What is the capital of France?", "Paris is the capital of France."],
    ["What is the capital of France?", "Berlin is in Germany."],
])

Specifications

Attribute	Value
Parameters	4B
Architecture	Qwen3ForSequenceClassification
Context	32,768 tokens
Languages	100+ (multilingual)
License	Apache 2.0

Use Cases

1. RAG pipelines — rerank retrieved chunks before passing to LLM
2. Search engines — improve document ranking quality
3. QA systems — score answer candidates for relevance
4. Semantic deduplication — score similarity for clustering

Abliteration

Like all Zen models, Zen Reranker is abliterated — refusal bias has been removed using directional ablation via hanzoai/remove-refusals.

Technique: Refusal in LLMs is mediated by a single direction — Arditi et al.

Model Family

Model	Parameters	Use Case
Zen Nano	0.6B	Edge AI
Zen Scribe	4B	Writing
Zen Pro	8B	Professional AI
Zen Reranker	4B	Retrieval
Zen Embedding	—	Embeddings

Citation

@misc{zen-reranker-2025,
  title={Zen Reranker: High-Performance Neural Reranking},
  author={Hanzo AI and Zoo Labs Foundation},
  year={2025},
  url={https://huggingface.co/zenlm/zen-reranker}
}

---

Part of the Zen model ecosystem by Hanzo AI (Techstars '17) and Zoo Labs Foundation.