BGE M3 pricing.
Prices checked
BGE M3 costs $0.02 per million input tokens from DigitalOcean Gradient; the cheapest price is $0.02 at DigitalOcean Gradient, checked .
DigitalOcean Gradient's own price
DigitalOcean Gradient
per million input tokens
80% below the typical price of the 14 embedding models we track.
Cheapest price
DigitalOcean Gradient
per million input tokens
Nobody sells it for less than DigitalOcean Gradient with a reason we can point at.
Every provider selling BGE M3
- 1
per million input tokens
$0.02/M tokens
Not ranked
Below the maker's own price with nothing published to account for it, from a catalog we could not read reliably, or a free tier, which is a limit rather than a price. Why
given away with limits
free
How we count
Prices are per million input tokens; an embedding model writes no output, so there is no blend. A price below DigitalOcean Gradient's own is listed as cheapest only when the provider says why. Beijing Academy of Artificial Intelligence trained BGE M3 and sells it on no catalog we read, so the price every row is measured against is DigitalOcean Gradient's.
- Same as maker
- Sold by someone else at the price the maker charges.
- Free tier
- Given away inside a quota. A limit is not a price, so it is never called cheapest.
Similar models
Embedding models from other labs, nearest in price first. Each percentage is against BGE M3's cheapest price, $0.02 per million input tokens.
Qwen3 Embedding 8B
33k context · open weights · 14 providers
DeepInfra$0.01/M tokens50% lesstext-embedding-3-small
8k context · closed weights · 9 providers
OpenAI$0.02/M tokensSame priceMistral Embed
8k context · closed weights · 3 providers
Mistral AI$0.10/M tokens400% moreGemini Embedding 2
8k context · closed weights · 6 providers
Google$0.20/M tokens900% more
Frequently asked
- Who sells BGE M3 cheapest?
- DigitalOcean Gradient, at $0.02 per million input tokens, checked 21 Oct 2026. Sold by someone else at the price the maker charges.
- Why is BGE M3 priced on input alone?
- An embedding model reads text and returns a vector, so there are no output tokens to bill. The price is per million tokens sent, which is why it is never compared with a chat model's blended price.
- Is BGE M3 open weights?
- Yes. Beijing Academy of Artificial Intelligence publishes the weights, so anyone may host BGE M3, which is why 1 providers sell it and why their prices differ.
- How does BGE M3 compare with other embedding models?
- At $0.02 per million input tokens, DigitalOcean Gradient's own price is 80% below the typical $0.10 across the 14 embedding models we track.
More from Beijing Academy of Artificial Intelligence
See also