Model reference · open weights
cross-encoder-mmarco-german-distilbert is an open-weight embedding model from ml6team. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Maker | ml6team |
|---|---|
| Type | Embedding models |
| Task | Reranker |
| Parameters (lead) | 135M |
| Context | 512 tokens |
| Runs with | sentence-transformers |
| Released | 2022-04-26 |
| Popularity | 1k downloads / month |
| Licence | Open weights |
About
This model is a fine-tuned cross-encoder on the MMARCO dataset which is the machine translated version of the MS MARCO dataset. As base model for the fine-tuning we use distilbert-base-multilingual-cased
Model input samples are tuples of the following format, either
assigned to 1 or assigned to 0.
The model was trained for 1 epoch.
The cross-encoder model can be used like this:
from sentence_transformers import CrossEncoder
model = CrossEncoder('model_name')
scores = model.predict([('Query 1', 'Paragraph 1'), ('Query 2', 'Paragraph 2')])
The model will predict scores for the pairs ('Query 1', 'Paragraph 1') and ('Query 2', 'Paragraph 2').
For more details on the usage of the cross-encoder models have a look into the Sentence-Transformers
Model evaluation was done on 2000 evaluation paragraphs of the dataset.
| Accuracy | F1-Score | Precision | Recall |
|---|---|---|---|
| 89.70 | 86.82 | 86.82 | 93.50 |
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys cross-encoder-mmarco-german-distilbert for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (cross-encoder-mmarco-german-distilbert 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":"cross-encoder-mmarco-german-distilbert","input":"text to embed"}'
Create an account — your API key is available in the console. 5M tokens/month currently included with every new account at launch.