Fireworks AI · Rate Limits
Fireworks Ai Rate Limits
Fireworks Ai Rate Limits is the rate-limit profile for Fireworks AI on the APIs.io network — every limit transcribed verbatim from the provider’s own documentation on 2026-10-07.
It captures 9 rate-limit definitions, including Total Prompt TPM — Small tier — 108M; Uncached Prompt TPM — Small tier — 27M; Generated TPM — Small tier — 1.08M.
Limit state is signalled in the X-Ratelimit-Limit-Tokens-Prompt, X-Ratelimit-Limit-Tokens-Cache-Adjusted-Prompt, and X-Ratelimit-Limit-Tokens-Generated response headers, and an exhausted limit returns HTTP 429.
Tagged areas include Artificial Intelligence, LLM, Inference, Multi-Modal, and Fine-Tuning.
9 Limits
Exhausted: HTTP 429
X-Ratelimit-Limit-Tokens-PromptX-Ratelimit-Limit-Tokens-Cache-Adjusted-PromptX-Ratelimit-Limit-Tokens-Generated
Transcribed from the provider's documentation on 2026-10-07 — docs.fireworks.ai/serverless/rate-limits.
View Source
Artificial IntelligenceLLMInferenceMulti-ModalFine-TuningGPU
Limits
Total Prompt TPM — Small tier accountSmall (< 600B total parameters)
108M
Uncached Prompt TPM — Small tier accountSmall (< 600B total parameters)
27M
Generated TPM — Small tier accountSmall (< 600B total parameters)
1.08M
Total Prompt TPM — Medium tier accountMedium (600B – < 1.6T)
43.2M
Uncached Prompt TPM — Medium tier accountMedium (600B – < 1.6T)
10.8M
Generated TPM — Medium tier accountMedium (600B – < 1.6T)
432k
Total Prompt TPM — Large tier accountLarge (≥ 1.6T)
21.6M
Uncached Prompt TPM — Large tier accountLarge (≥ 1.6T)
5.4M
Generated TPM — Large tier accountLarge (≥ 1.6T)
216k
Sources
Work with this as data
Every rate limit here is available over the APIs.io API and to AI agents over MCP.