> ## Documentation Index
> Fetch the complete documentation index at: https://docs.apiyi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Gemini 3.1 Flash Lite goes GA

> Google announced GA on May 8 (UTC+8). The model name drops -preview and becomes gemini-3.1-flash-lite — contract is now production-stable. Output 64% faster than 2.5 Flash with 2.5× faster TTFT, GPQA Diamond 86.9% / MMMU Pro 76.8%. Listed at \$0.25/\$1.50 per 1M tokens, identical to Google's rates; stackable top-up promo brings effective price down to ~79–85% of list.

**2026/5/9 16:53 (UTC+8)** · New Model · Google

🚀 **Gemini 3.1 Flash Lite goes GA** — Google announced GA on **May 8 (UTC+8)**. The model name drops `-preview` and becomes `gemini-3.1-flash-lite`, with the API contract frozen and production-ready. Output is **64% faster** than 2.5 Flash (381.9 vs 232.3 t/s) and **2.5× faster** on TTFT; GPQA Diamond **86.9%** / MMMU Pro **76.8%**, 1M context + full multimodal input.

💰 **Pricing identical to Google's rates**: \$0.250 / \$1.500 per 1M tokens (cached input \~10% of list). APIYi's official direct-connection channel is live in sync; stacking the top-up promo brings effective cost to **\~79–85% of list**.

📖 Launch notes: [/en/news/gemini-3-1-flash-lite-launch](/en/news/gemini-3-1-flash-lite-launch)

***

← [Back to Live Updates](/en/live) · 📚 [Monthly Archive](/en/live/archive)