Jina Embedding v1 Tiny English
Jina Embedding v1 Tiny is an English sentence embedding model with 14 million parameters and 312-dimensional output. Jina's Finetuner team trained the family for retrieval and textual similarity.
How the model works
The release uses Linnaeus-Clean, a collection of 380 million sentence pairs selected from a larger 1.6-billion-pair source. It includes query-document pairs across several domains. The weights use Apache 2.0 and can run with Sentence Transformers or Jina's historical Finetuner tooling.Jina Embedding v1 Tiny English model details
Use and access
Tiny is the smallest published v1 size. Jina reports CPU and GPU timing for short sentences in its card, with the hardware specified. Its low parameter count is a resource choice, not evidence of equivalent retrieval quality to the larger checkpoints.Jina Embedding v1 Tiny English model details
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | Downloadable |
| Model identifiers | jina-embedding-t-en-v1jinaai/jina-embedding-t-en-v1 |
| Parameters | 14 millionJina Embedding v1 Tiny English model details |
| Embedding dimensions | 312Jina Embedding v1 Tiny English model details |
| Weight license | Apache 2.0Jina Embedding v1 Tiny English model details |
Related models
- Jina Embeddings v2 Small English
Jina AI ยท Jina Embeddings v2
A 33-million-parameter English encoder with long-document input.
Launch date Downloadable