Our take
Fireworks is a performance-focused host for downloadable models with a strong verified trust posture. It has an EU endpoint, an independent security attestation, and a commitment not to train on prompts.
Use this for production downloadable-model inference where compliance posture matters, or for teams standardising on one host for several open-weights flagships. Skip it if price is the only criterion. It is not the cheapest option for the models it hosts.
- EU endpoint, SOC 2 and no training on prompts — all three verified rather than claimed.
- Current open-weights flagship coverage including GLM 5.2, Kimi K2.7 Code and DeepSeek V4 Flash.
- Not the price leader: GLM 5.2 costs noticeably more here than at the cheapest hosts.
Point your tools here
We have not recorded what a router needs for Fireworks AI yet — no base URL and no docs link on file. That is our gap, not a sign Fireworks AI has no API; its own documentation is the place to look until we close it.
Privacy & data handling
These answers cover requests sent to Fireworks AI through OpenRouter, as OpenRouter records them, last read 3 hours ago. For Fireworks AI's own API we hold no answer: check its terms. SOC 2 and the data processing agreement below are hand-checked, not part of that sync.
We have not dated this attestation or linked the report yet.
Prompt retention: none.