There is no per-unit API price: price not verified.
#31 of 200 in Embeddings. Strongest at: Embeddings & Retrieval.
On MTEB Portuguese v1 · Multilabel Classification · Portuguese, Snowflake Arctic Embed L scores 23.2 vs 23.2 — about the same.
Released 2024-04-12. 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.
Snowflake Arctic Embed L is an open-weights embeddings & retrieval model from Snowflake, released Apr 2024.
Gemini Embedding 001 (better), Giga Embeddings Instruct 3B 0826 (open weights).
BetterGemini Embedding 001unstated+21%—$0.15aOpenGiga Embeddings Instruct 3B 0826unstated+48%——Memory (4-bit): 0.2 GB → 1.9 GB
BetterHarrier OSS v1 27BOpen+64%——Memory (4-bit): 0.2 GB → 16 GBaBetterGiga Embeddings Instruct 10B A1.8B 0826unstatedOpen+46%——Memory (4-bit): 0.2 GB → 6.3 GB
BetterKaLM Embedding Gemma3 12B 2511Open+36%——Memory (4-bit): 0.2 GB → 7.1 GB
Llama Embed Nemotron 8BOpen+25%——Memory (4-bit): 0.2 GB → 4.5 GB