# Embedding Models

Self-hosted mirror of permissively-licensed retrieval models used by [xberg](https://github.com/xberg-io/xberg). Weights are unmodified; every file is sha256-pinned in the xberg source tree (`crates/xberg/src/embeddings/presets.sha256sum`) and verified at download.

All models are Apache-2.0 or MIT licensed. Attribution to each upstream author below.

| Model | Upstream | License | Notes |
|---|---|---|---|
| `all-MiniLM-L6-v2` | [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) | Apache-2.0 | dense embedding, 384d |
| `bge-base-en-v1.5` | [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) | MIT | dense embedding, 768d |
| `bge-large-en-v1.5` | [BAAI/bge-large-en-v1.5](https://huggingface.co/BAAI/bge-large-en-v1.5) | MIT | dense embedding, 1024d |
| `multilingual-e5-base` | [intfloat/multilingual-e5-base](https://huggingface.co/intfloat/multilingual-e5-base) | MIT | multilingual dense, 768d |
| `gte-modernbert-base` | [Alibaba-NLP/gte-modernbert-base](https://huggingface.co/Alibaba-NLP/gte-modernbert-base) | Apache-2.0 | dense embedding, 768d, 8k ctx (default) |
| `potion-base-8m` | [minishlab/potion-base-8M](https://huggingface.co/minishlab/potion-base-8M) | MIT | static model2vec, 256d, no ORT |
| `arctic-embed-m-v2.0` | [Snowflake/snowflake-arctic-embed-m-v2.0](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v2.0) | Apache-2.0 | asymmetric multilingual dense, 768d |
| `qwen3-embedding-0.6b` | [Qwen/Qwen3-Embedding-0.6B](https://huggingface.co/Qwen/Qwen3-Embedding-0.6B) | Apache-2.0 | decoder dense, 1024d, last-token, 32k ctx |