Tracked providers
| Provider | Input /1M tokens | Output /1M tokens | Provenance |
|---|---|---|---|
| $0.750 | $3.750 | vendor reported | |
| oneinfer | $0.825 | $4.125 | vendor reported |
| $0.500 | $6.000 | vendor reported | |
| $0.500 | $9.000 | vendor reported | |
| fal | Not published | Not published | vendor reported |
| fal | Not published | Not published | vendor reported |
What is Gemini 3.8 Flash?
Gemini 3.8 Flash is Google's proprietary fast reasoning model released on 2 September 2026. It accepts text, image, audio, video and PDF, returns text, and is served through OneInfer under the model identifier google/gemini-3.8-flash.
Why route count is not provider diversity
Weights are closed. Every aggregator, gateway and proxy resolves upstream to Google. Outages, policy changes, rate-limit behaviour and model updates retain a single-owner dependency. There is currently one first-party provider benchmarked by Artificial Analysis: Google AI Studio.
Ready to test the workflow?
Gemini 3.8 Flash Cyber is not in any provider table
The Cyber variant is restricted to Google's Fairwind Program. It is not listed by AIHubMix, LiteLLM, Vercel or LLM Gateway. If a third-party route claims to sell Gemini 3.8 Flash Cyber, treat the claim with suspicion; verify with Google directly.