| --- |
| license: apache-2.0 |
| base_model: nomic-ai/nomic-embed-text-v1.5 |
| library_name: basert |
| pipeline_tag: feature-extraction |
| tags: |
| - basert |
| - apple-silicon |
| - quantized |
| - sentence-similarity |
| --- |
| |
| # nomic-embed-text-v1.5 |
|
|
| [BaseRT](https://github.com/basecompute/baseRT) `.base` builds of [`nomic-ai/nomic-embed-text-v1.5`](https://huggingface.co/nomic-ai/nomic-embed-text-v1.5), a 768-dimension text-embedding model, for fast local inference on Apple Silicon. |
|
|
| ## Files |
|
|
| | File | Precision | Size | |
| |------|-----------|------| |
| | `nomic-embed-text-v1.5-Q4.base` | 4-bit | 78 MB | |
| | `nomic-embed-text-v1.5-Q8.base` | 8-bit | 142 MB | |
|
|
| ## Usage |
|
|
| ```bash |
| curl -LsSf https://basecompute.co/install.sh | sh |
| basert serve --model nomic-embed-text-v1.5-Q8.base |
| ``` |
|
|
| `POST /v1/embeddings` returns 768-dimension, L2-normalized vectors. |
|
|
| For retrieval quality, the 8-bit build is recommended. |
|
|
| Released under the apache-2.0 license, inherited from the base model. |
|
|
