Token (Meta Model API)
Catalog row in Vendor License Metrics · Cited
- Unit
- Token
- Aliases
- Tokens; per 1M tokens
- Definition
- Pricing and rate limits: Meta Model API "bills text models per token". Prices are stated per 1M tokens for cached input, input and output.
- Counting rules
- Input, cached input and output tokens are priced separately per tier. Cached-input rate applies to tokens that match a cached prefix (cached_tokens in the response). Tokens that Meta injects as steering context are not billed. No long-context premium. Rate limits are per team, not per API key.
- Floors / minimums
- No minimums or upfront commitment.
- Applies to products
- Muse Spark
- Product category