There is no per-unit API price: self-host only.
There is no free API. The weights are open, so you can run it yourself.
#100 of 200 in Embeddings.
On MTEB multilingual v2 · Reranking · multilingual (22 languages), Nomic Embed Text v1 Ablated scores 45 vs 45.6 — about the same.
Released 2024-01-15. 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.
Nomic Embed Text v1 Ablated is an open-weights embeddings & retrieval model from nomic-ai, released Jan 2024.
EmbeddingGemma 300M (better), F2LLM v2 8B (open weights).
BetterEmbeddingGemma 300MunstatedOpen+40%—$0.002Max input: 8K tokens → 2K tokens+1cOpenF2LLM v2 8Bunstated+56%——Memory (4-bit): 0.1 GB → 4.5 GB
BetterBge Base En v1.5unstatedOpen+6%—$0.005Max input: 8K tokens → 1K tokensiBetterE5 Base v2unstatedOpen+5%—$0.005Max input: 8K tokens → 1K tokens
BetterQwen 3 Embedding 8BunstatedOpen+46%—$0.01Memory (4-bit): 0.1 GB → 4.5 GB
Llama Embed Nemotron 8BOpen+50%——Memory (4-bit): 0.1 GB → 4.5 GBOOcten Embedding 8BOpen+50%——Memory (4-bit): 0.1 GB → 4.5 GB