Model reference · open weights
ettin-reranker is an open-weight embedding model from cross-encoder, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
ettin-reranker-68m-v1 This is a Cross Encoder model finetuned from jhu-clsp/ettin-encoder-68m on the cross-encoder/ettin-reranker-v1-data dataset using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search. See the release blogpost for details on the training recipe, evaluation results, and speed benchmarks against other public rerankers. The Evaluation section below also has the headline numbers. Model Details Model Description - Model Type: Cross Encoder - Base model: jhu-clsp/ettin-encoder-68m <!-- at revision ac19ae4bc51093b31c475665ac872a936d056cc2 -- - Maximum Sequence Length: 7999 tokens - Number of Output Labels: 1 label - Supported Modality: Text - Training Dataset: cross-encoder/ettin-reranker-v1-data - Language: en - License: apache-2.0 Model Sources - Documentation: Sentence Transformers Documentation - Documentation: Cross Encoder Documentation - Repository: Sentence Transformers on GitHub - Hugging Face: Cross Encoders on Hugging Face Full Model Architecture Usage Direct Usage (Sentence Transformers) First install the Sentence Transformers library: Then you can load this model and run inference. Direct Usage (Transformers) -- Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. -- Out-of-Scope Use List how the model may foreseeably be misused and address what users ought not to do with the model. -- Evaluation MTEB(eng, v2) Retrieval Each model in the ettin-reranker-v1 family was evaluated on the full MTEB(eng, v2) Retrieval benchmark (10 tasks, top-100 reranked) using MTEB's two-stage reranking flow, pairing each reranker with six embedding models that span the speed/quality spectrum. The dashed retriever-only line in each chart below is the headline number to beat. Anything below it means the reranker actively hurts the pipeline on average: Mean NDCG@10 over the 6 embedder pairings, sorted by MTEB. The released ettin-reranker-v1 family is in bold, and the teacher mixedbread-ai/mxbai-rerank-large-v2 is <uunderlined</u. See the release blogpost for the full analysis and per-model commentary. Speed All six released models were benchmarke
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | cross-encoder |
|---|---|
| Type | Embedding models |
| Parameters (lead) | 68M |
| Context | 7,999 tokens |
| Variants | 3 |
| Runs with | sentence-transformers |
| Based on | jhu-clsp/ettin-encoder-68m |
| Released | 2026-05-15 |
| Popularity | 17k downloads / month |
| Likes | 5 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
Benchmarks
As published on the model card — the maker's own numbers, not measured by AxForge.
| Task | Dataset | Metric | Score |
|---|---|---|---|
| Cross Encoder Reranking | NanoMSMARCO R100 | Map | 0.617 |
| Cross Encoder Reranking | NanoMSMARCO R100 | Mrr@10 | 0.613 |
| Cross Encoder Reranking | NanoMSMARCO R100 | Ndcg@10 | 0.683 |
| Cross Encoder Reranking | NanoNFCorpus R100 | Map | 0.372 |
| Cross Encoder Reranking | NanoNFCorpus R100 | Mrr@10 | 0.562 |
| Cross Encoder Reranking | NanoNFCorpus R100 | Ndcg@10 | 0.407 |
| Cross Encoder Reranking | NanoNQ R100 | Map | 0.722 |
| Cross Encoder Reranking | NanoNQ R100 | Mrr@10 | 0.753 |
| Cross Encoder Reranking | NanoNQ R100 | Ndcg@10 | 0.775 |
| Cross Encoder Reranking | NanoFiQA2018 R100 | Map | 0.538 |
| Cross Encoder Reranking | NanoFiQA2018 R100 | Mrr@10 | 0.652 |
| Cross Encoder Reranking | NanoFiQA2018 R100 | Ndcg@10 | 0.591 |
| Cross Encoder Reranking | NanoTouche2020 R100 | Map | 0.477 |
| Cross Encoder Reranking | NanoTouche2020 R100 | Mrr@10 | 0.826 |
| Cross Encoder Reranking | NanoTouche2020 R100 | Ndcg@10 | 0.563 |
| Cross Encoder Reranking | NanoSciFact R100 | Map | 0.702 |
| Cross Encoder Reranking | NanoSciFact R100 | Mrr@10 | 0.7 |
| Cross Encoder Reranking | NanoSciFact R100 | Ndcg@10 | 0.746 |
| Cross Encoder Reranking | NanoHotpotQA R100 | Map | 0.932 |
| Cross Encoder Reranking | NanoHotpotQA R100 | Mrr@10 | 0.99 |
| Cross Encoder Reranking | NanoHotpotQA R100 | Ndcg@10 | 0.958 |
| Cross Encoder Reranking | NanoArguAna R100 | Map | 0.574 |
| Cross Encoder Reranking | NanoArguAna R100 | Mrr@10 | 0.575 |
| Cross Encoder Reranking | NanoArguAna R100 | Ndcg@10 | 0.688 |
Using it via the API
Once AxForge deploys ettin-reranker for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (ettin-reranker below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/embeddings \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"ettin-reranker","input":"text to embed"}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗
Explore