NezhaGateNezhaGate
AnnouncementsAll updates →
NoticeWan 3.0: Standard 480P / 720P / 1080P available; Prime paused NewKling 4.0 is live: reference images, videos and audio, 3-30 seconds, billed per second at 720P / 1080P NoticeFrom October 29: top-up bonuses by payment method and a $1.40 refund fee
简体中文English ✓Español한국어日本語Русский
Log in Get API Key
Chat ● Operational Service status →

Gemini 3.8 Flash

⧉
gemini-3.8-flash

Gemini 3.8 Flash is the latest Flash generation from Google (released September 2026), built for long-running coding, agent workflows and complex reasoning: 1M-token context, up to 64K output, image input. It uses a dynamic thinking budget -- quick answers for simple questions, more thinking for hard ones. Fully OpenAI-API compatible: set model to gemini-3.8-flash.

ChatLatest modelAdaptive thinking1M contextOpenAI compatible
Get API Key → API Docs

Live Test · Playground

Try out Gemini 3.8 Flash right here (available after login).

Input

Advanced
Web search
Memory · multi-turn

Conversation

Start a conversationType a message below to begin

About Gemini 3.8 Flash

Gemini 3.8 Flash is the latest Flash generation from Google (released September 2026), built for long-running coding, agent workflows and complex reasoning: 1M-token context, up to 64K output, image input, and a dynamic thinking budget that answers simple questions fast and thinks longer on hard ones. Fully OpenAI-API compatible: set model to gemini-3.8-flash. Priced below the Google list price on both input and output.

Use cases

Long-running coding and agent workflows

A 1M context holds a whole repository plus long conversation history, ideal for agents with many tool calls.

Batches of uneven difficulty

The dynamic thinking budget returns simple requests quickly and thinks more on hard ones, with no manual tiering.

Mixed image and text input

Image input is supported, so screenshot understanding and chart questions use the same endpoint.

How to choose

Pick 3.8 for the latest generation; stay on gemini-3.7-flash (same price) if your prompts are tuned for it; use the four gemini-3.6-flash tiers when you want manual control of the thinking budget.

FAQ

How does it differ from gemini-3.7-flash?
3.8 is the latest generation and 3.7 the previous one; same price, same dynamic thinking budget, with 3.8 aimed at longer-running coding and agent tasks.
Is it cheaper than the official price?
Yes. Both input and output are priced below the Google list price ($0.75 / $3.75 per 1M); the pricing page shows the comparison line by line.
Context and output limits?
1M-token context and up to 64K output; system prompts and other overhead count toward the window.
Am I charged for failed calls?
No. Only calls that return successfully are settled on actual usage; upstream errors and timeouts never touch your balance.

Related Models

Explore other models you can integrate.

View all →