$0.05 per 1M tokens.
There is no free API; it costs $0.05 per 1M tokens. The weights are open, so you can run it yourself.
#169 of 200 in Embeddings.
On JMTEB v2 · STS · Japanese, Jina Embeddings v2 Base En scores 27.7 vs 28.6 — about the same.
Released 2023-09-27. It is the current version in its line.
Yes, the weights are open (Apache-2.0 licence).
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
Jina Embeddings v2 Base En is an open-weights embeddings & retrieval model from Jina AI, released Sep 2023.
EmbeddingGemma 300M (cheaper), Ja Embed Sts (open weights).
CheaperEmbeddingGemma 300MunstatedOpen×3.1-96%$0.002Max input: 8K tokens → 2K tokens+2NOpenJa Embed Sts×3.1——Max input: 8K tokens → 1K tokens+1iCheaperE5 Mistral 7B InstructunstatedOpen×3-55%$0.022Memory (4-bit): 0.1 GB → 4.3 GBiCheaperMultilingual E5 Large InstructunstatedOpen×3-80%$0.01Max input: 8K tokens → 1K tokens+1iCheaperMultilingual E5 LargeunstatedOpen×2.9-80%$0.01Max input: 8K tokens → 1K tokens+1
Bge M3unstatedOpen×2.9-80%$0.01Languages: not verified+1
Harrier OSS v1 0.6BunstatedOpen×3.1——Memory (4-bit): 0.1 GB → 0.4 GBcF2LLM v2 4BunstatedOpen×3.1——Memory (4-bit): 0.1 GB → 2.4 GB