Gemini 3.8 Flash Is Now Available

Google has released Gemini 3.8 Flash, calling it their most intelligent Flash model yet, and it is generally available from day one. It is now the recommended Google model in AI Data Stream.

What Changed for You

Nothing you need to do. Gemini 3.8 Flash is already in the model picker as the default choice for Google.

• Gemini 3.7 Flash stays available, now labelled “previous gen”. If you have tuned prompts or reports against it, nothing changes until you decide to switch.
• Gemini 2.5 Flash has been retired. Anyone still using it is migrated to Gemini 3.8 Flash automatically. Note that this is a price step up: 2.5 Flash was $0.30 / $2.50 per million tokens, and 3.8 Flash is $0.75 / $3.75. If you would rather stay on a cheaper model, pick Gemini 2.5 Flash Lite or 3.5 Flash Lite from the picker.
• Gemini 2.5 Flash Lite is not being migrated. It stays as the cheapest option, so nobody’s costs change without them choosing it.

Your API key, settings, and conversations all remain the same.

Google also announced Gemini 3.8 Flash Cyber, a security-focused variant. It is only available to vetted organisations through Google’s Fairwind Program, so it is not part of the AI Data Stream lineup.

Pricing

As always, AI Data Stream is bring-your-own-key: you are billed by Google directly at their published rates.

• Gemini 3.8 Flash has the same introductory pricing as 3.7 Flash: $0.75 / $3.75 per million input / output tokens through 31 December 2026. From 1 January 2027 the standard price is $1.50 / $7.50.
• Gemini 3.7 Flash stays at the same $0.75 / $3.75 introductory price, then $1.50 / $7.50.
• Gemini 3.1 Pro is $2.00 / $12.00 per million tokens for prompts up to 200k tokens ($4.00 / $18.00 above that).
• Gemini 3.5 Flash Lite is $0.30 / $2.50 per million tokens.
• Gemini 2.5 Flash Lite remains the budget choice at $0.10 / $0.40 per million tokens.

What Is New in Gemini 3.8 Flash

Gemini 3.8 Flash keeps the 1M-token context window and tunable thinking levels of 3.7 Flash, and Google reports significant gains in agentic tasks and multi-step reasoning, often approaching frontier-model performance. That matters for AI Data Stream because every analysis is a chain of tool calls against your connected data sources, and the more reliably the model plans and follows through on those calls, the better the answers. Google Search grounding is supported, as it is across our whole Gemini lineup.

The Full Gemini Lineup

With this update, the complete Google model lineup in AI Data Stream is:

• Gemini 3.8 Flash - Best balance, the recommended default
• Gemini 3.1 Pro - Most advanced, for the most demanding analysis
• Gemini 2.5 Pro - Previous-generation reasoning model
• Gemini 3.7 Flash - Previous-generation workhorse
• Gemini 3.5 Flash Lite - Fastest
• Gemini 2.5 Flash Lite - Cheapest

Switching Models

If you want to try Gemini 3.8 Flash or switch between models:

1. Go to Settings > AI Providers
2. Click your Google provider
3. Select your preferred model from the dropdown
4. Save

Or simply select it from the model picker when starting a new conversation.

Questions?

If you have any questions about the transition, contact us or check our AI providers documentation.