$0.01 per 1M tokens.
There is no free API; the cheapest paid price is $0.01 per 1M input tokens. The weights are open, so you can run it yourself.
#3 of 200 in Embeddings. Strongest at: Embeddings & Retrieval.
On MTEB multilingual v2 · Bitext Mining · multilingual (1026 languages), Qwen 3 Embedding 8B scores 80.9 vs 81.5 — about the same.
Through the Alibaba (Qwen) API.
Released 2025-06-05. 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.
Qwen 3 Embedding 8B is an open-weights embeddings & retrieval model from Alibaba (Qwen), released Jun 2025.
Harrier OSS v1 27B (better), KaLM Embedding Gemma3 12B 2511 (open weights).