$0.002 per 1M tokens.
There is no free API; the cheapest paid price is $0.002 per 1M input tokens. The weights are open, so you can run it yourself.
#53 of 200 in Embeddings.
On MTEB Indic v1 · Reranking · multilingual (2 languages), EmbeddingGemma 300M scores 88.8 vs 88.8 — about the same. It is 97% cheaper.
Through the Google API.
Released 2025-09-04. It is the current version in its line.
Yes, the weights are open (gemma licence).
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
EmbeddingGemma 300M is an open-weights embeddings & retrieval model from Google, released Sep 2025.
Octen Embedding 8B (open weights).
OpenLlama Embed Nemotron 8Bsame——Memory (4-bit): 0.2 GB → 4.5 GBaOpenGiga Embeddings Instruct 10B A1.8B 0826unstatedsame——Memory (4-bit): 0.2 GB → 6.3 GBLOpenFlowVec v1same——Max input: 2K tokens → 1K tokens+1cOpenF2LLM v2 8Bunstatedsame——Memory (4-bit): 0.2 GB → 4.5 GB