Setting thinking_level and legacy thinking_budget together returns HTTP 400; use only one. Source ↗
$2 per 1M input tokens and $12 per 1M output tokens (cheapest host: 302.AI).
There is no free API; the cheapest paid price is $2 per 1M input tokens.
Through the Google API.
1M tokens.
Released 2025-11-18. It has been superseded by Gemini 3.1 Pro Preview.
No, Gemini 3 Pro Preview is a closed model available through an API. The closest open-weights option here is MiMo v2.6 Pro.
It accepts text and images, PDF files, audio, video as input.
Gemini 3 Pro Preview is a closed language model from Google, released Nov 2025.
Claude Sonnet 5.5 (cheaper), GPT-6 Luna (better), MiMo v2.6 Pro (open weights).