Google's Gemini 3.8 Flash is now available on OpenRouter, served by both Google Vertex and Google AI Studio with automatic failover between them.

Flash-class pricing with a 1M token context window is reachable through one OpenRouter API key without Google Cloud credentials, with routing across two providers and automatic failover.

Google Gemini 3.8 Flash Now Available on OpenRouter

Q: What changed?

Pricing holds at $0.75 per million input tokens and $3.75 per million output, with a 1,048,576-token context window, 65,536 max output tokens, a batch variant at half price, and cache reads at $0.075 per million tokens.

What shipped

Google's Gemini 3.8 Flash is now available on OpenRouter (added September 2, 2026), served by both Google Vertex and Google AI Studio with automatic failover between them.

What changed

Pricing holds at the 3.7 Flash level — $0.75 per million input tokens, $3.75 per million output — while the model has been rebuilt for long-horizon work: a 1,048,576-token context window, 65,536 max output tokens, and a focus on software engineering, autonomous agents, and enterprise workflows. A :batch variant runs at half price ($0.375/$1.875 per million), and cache reads drop to $0.075 per million. OpenRouter routes requests across two providers with pinning available, and the usual gateway features (logging, fallbacks) apply through a single API key.

Why a builder cares

Flash-class latency and pricing with frontier-grade context — reachable through one API key, no Google Cloud credential setup, and with provider failover on top. Swapping google/gemini-3.7-flash for google/gemini-3.8-flash is a one-line change.

What shipped

What changed

Why a builder cares

Related reading