Gemini 3.8 Flash now available on AI Gateway
Gemini 3.8 Flash hits AI Gateway at 50% off—half a million devs just got faster, cheaper multi-step reasoning for agents.
Why it matters
Google's latest Flash model is now accessible through Vercel's AI Gateway with significant pricing incentives and native support for agentic workflows. Practitioners building with Claude Code, Cursor, and other agent frameworks can plug it in immediately.
The key facts
17 to knowGemini 3.8 Flash launched on Vercel AI Gateway
50% discount through December 31st
1M token context window
Supports text, image, PDF, video input
Max output: 65,536 tokens
Improved performance on software engineering, agent work, multi-step reasoning
Thinking enabled by default
Tool calling and web search supported
Native integration with Claude Code, OpenCode, Cursor, Pi agents
AI Gateway charges zero platform fee on inference (including BYOK)
Gemini 3.8 Flash available on Vercel AI Gateway
50% pricing discount through December 31, 2026
Supports text, image, PDF, video input; text output
65,536 token max output
Improvements in software engineering, agent work, multi-step reasoning vs. prior Flash
Integrates with Claude Code, OpenCode, Cursor, Pi agents
AI Gateway reflects provider pricing with zero platform markup on inference
Go to the source
Vercel Blogvercel.com
Publisher excerpt: is now available on AI Gateway.Gemini 3.8 Flash from Google The model is 50% off through December 31st. It has a 1M token context window, accepts text, image, PDF, and video input, returns text, and supports tool calling and web search. Maximum output is 65,536 tokens. Gemini 3.8 Flash improves on…