hs-rerank-1

hs-rerank-1 is a multilingual cross-encoder reranker built for Hindsight. It reads a query and a document together and returns a relevance score.

Highlights

• Multilingual: English, Spanish, French, German, Portuguese, Japanese, Korean and Chinese (Simplified and Traditional).
• Long inputs: up to 8,192 tokens per query and document pair.
• Small: 114M parameters. 98M of them are the token embedding table, so each forward pass runs a 15M-parameter transformer.
• CPU and GPU: ships an int8 ONNX export for CPU, and safetensors for GPU in FP16.
• Serves on TEI: CPU and GPU images.

Model overview

Model type	Cross-encoder reranker
Base model	jhu-clsp/mmBERT-small
Parameters	114M
Languages	en, es, fr, de, pt, ja, ko, zh
Context length	8,192 tokens
Output	One relevance score per query and document pair
Formats	safetensors stored in FP16 (FP32 copy on the fp32 branch), ONNX (onnx/model.onnx, int8 vocabulary table)

Evaluation

Usage

Sentence Transformers

pip install -U sentence-transformers

from sentence_transformers import CrossEncoder

model = CrossEncoder("hindsight-ai/hs-rerank-1")

query = "¿Qué app de presupuesto estoy creando?"
documents = [
    "El usuario quiere tooltips en su app de presupuesto.",
    "El usuario corre los domingos por la mañana.",
]

print(model.rank(query, documents))
# [{'corpus_id': 0, 'score': 0.9647}, {'corpus_id': 1, 'score': 0.0003}]

predict and rank return scores between 0 and 1. Pass activation_fn=torch.nn.Identity() for raw logits.

Text Embeddings Inference

CPU (uses onnx/model.onnx):

docker run -p 8080:80 \
  ghcr.io/huggingface/text-embeddings-inference:cpu-1.9.3 \
  --model-id hindsight-ai/hs-rerank-1 --auto-truncate \
  --max-client-batch-size 128

GPU:

docker run --gpus all -p 8080:80 \
  ghcr.io/huggingface/text-embeddings-inference:89-1.9.3 \
  --model-id hindsight-ai/hs-rerank-1 --auto-truncate \
  --max-client-batch-size 128

Use the GPU image tag for your GPU: 89-1.9.3 for L4 and other Ada GPUs, 1.9.3 for A100 and H100.

curl 127.0.0.1:8080/rerank \
  -H 'Content-Type: application/json' \
  -d '{
    "query": "¿Qué app de presupuesto estoy creando?",
    "texts": [
      "El usuario quiere tooltips en su app de presupuesto.",
      "El usuario corre los domingos por la mañana."
    ]
  }'

Hindsight

HINDSIGHT_API_RERANKER_PROVIDER=local
HINDSIGHT_API_RERANKER_LOCAL_MODEL=hindsight-ai/hs-rerank-1

Or point Hindsight at a TEI server:

HINDSIGHT_API_RERANKER_PROVIDER=tei
HINDSIGHT_API_RERANKER_TEI_URL=http://localhost:8080

Hindsight Inference

hs-rerank-1 is also served at https://inference.hindsight.vectorize.io. To use it from Hindsight:

HINDSIGHT_API_RERANKER_PROVIDER=hindsight
HINDSIGHT_API_RERANKER_HINDSIGHT_API_KEY=<your Hindsight Inference API key>

License

MIT

Citation

@misc{hindsightai2026hsrerank1,
  title  = {hs-rerank-1},
  author = {{Hindsight AI}},
  year   = {2026},
  url    = {https://huggingface.co/hindsight-ai/hs-rerank-1}
}