File size: 3,457 Bytes
c37cd6e e00a879 c37cd6e 9aea531 c37cd6e 9aea531 c37cd6e c11d8aa c37cd6e 9aea531 c37cd6e 3993464 9aea531 c37cd6e 8a3b707 811b39e 8a3b707 811b39e 8a3b707 9aea531 c37cd6e 9aea531 c37cd6e 48fac2a c37cd6e 9b51836 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 | ---
license: cc-by-nc-4.0
language:
- en
base_model:
- Qwen/Qwen3-4B
pipeline_tag: text-ranking
tags:
- finance
- legal
- code
- stem
- medical
library_name: sentence-transformers
---
<img src="https://i.imgur.com/oxvhvQu.png"/>
# Releasing zeroentropy/zerank-1
In search engines, [rerankers are crucial](https://www.zeroentropy.dev/blog/what-is-a-reranker-and-do-i-need-one) for improving the accuracy of your retrieval system.
However, SOTA rerankers are closed-source and proprietary. At ZeroEntropy, we've trained a SOTA reranker outperforming closed-source competitors, and we're launching our model here on HuggingFace.
This reranker [outperforms proprietary rerankers](https://huggingface.co/zeroentropy/zerank-1#evaluations) such as `cohere-rerank-v3.5` and `Salesforce/LlamaRank-v1` across a wide variety of domains, including finance, legal, code, STEM, medical, and conversational data.
At ZeroEntropy we've developed an innovative multi-stage pipeline that models query-document relevance scores as adjusted [Elo ratings](https://en.wikipedia.org/wiki/Elo_rating_system). See our Technical Report (Coming soon!) for more details.
Since we're a small company, this model is only released under a non-commercial license. If you'd like a commercial license, please contact us at founders@zeroentropy.dev and we'll get you a license ASAP.
For this model's smaller twin, see [zerank-1-small](https://huggingface.co/zeroentropy/zerank-1-small), which we've fully open-sourced under an Apache 2.0 License.
## How to Use
```python
from sentence_transformers import CrossEncoder
model = CrossEncoder("zeroentropy/zerank-1", trust_remote_code=True)
query_documents = [
("What is 2+2?", "4"),
("What is 2+2?", "The answer is definitely 1 million"),
]
scores = model.predict(query_documents)
print(scores)
```
The model can also be inferenced using ZeroEntropy's [/models/rerank](https://docs.zeroentropy.dev/api-reference/models/rerank) endpoint.
## Evaluations
NDCG@10 scores between `zerank-1` and competing closed-source proprietary rerankers. Since we are evaluating rerankers, OpenAI's `text-embedding-3-small` is used as an initial retriever for the Top 100 candidate documents.
| Task | Embedding | cohere-rerank-v3.5 | Salesforce/Llama-rank-v1 | zerank-1-small | **zerank-1** |
|----------------|-----------|--------------------|--------------------------|----------------|--------------|
| Code | 0.678 | 0.724 | 0.694 | 0.730 | **0.754** |
| Conversational | 0.250 | 0.571 | 0.484 | 0.556 | **0.596** |
| Finance | 0.839 | 0.824 | 0.828 | 0.861 | **0.894** |
| Legal | 0.703 | 0.804 | 0.767 | 0.817 | **0.821** |
| Medical | 0.619 | 0.750 | 0.719 | 0.773 | **0.796** |
| STEM | 0.401 | 0.510 | 0.595 | 0.680 | **0.694** |
Comparing BM25 and Hybrid Search without and with zerank-1:
<img src="https://cdn-uploads.huggingface.co/production/uploads/67776f9dcd9c9435499eafc8/2GPVHFrI39FspnSNklhsM.png" alt="Description" width="400"/> <img src="https://cdn-uploads.huggingface.co/production/uploads/67776f9dcd9c9435499eafc8/dwYo2D7hoL8QiE8u3yqr9.png" alt="Description" width="400"/> |