# Gemini 3.6 Flash

Gemini 3.6 Flash improves coding, agentic tasks, and web development while spending fewer output tokens and fewer model calls than Gemini 3.5 Flash, so a multi-step agent run finishes in fewer steps.

- **Model ID:** `google/gemini-3.6-flash`
- **Type:** chat
- **Providers:** vertex, google
- **Context window:** 1,000,000
- **Maximum output tokens:** 64,000
- **Pricing:** $0.75/1M input tokens, $3.75/1M output tokens
- **Canonical page:** https://vercel.com/ai-gateway/models/gemini-3.6-flash

## Supported parameters

Detailed capability metadata has not been reported for this model.

## Example

```ts
import { streamText } from 'ai'

const result = streamText({
  model: 'google/gemini-3.6-flash',
  prompt: 'Why is the sky blue?'
})
```

## About

Gemini 3.6 Flash is Google's Flash-tier workhorse, released on July 21, 2026, and it builds directly on Gemini 3.5 Flash. Quality rises across coding, knowledge work, and multimodal tasks while token consumption falls. On the Artificial Analysis Index, Gemini 3.6 Flash consumes fewer output tokens than Gemini 3.5 Flash, and it completes multi-step workflows with fewer reasoning steps and fewer tool calls. For agent workloads, that compounds: the saving lands on every step of a run, not on a single call.

Coding is where the gains are easiest to measure. Gemini 3.6 Flash scores 49% on DeepSWE against 37% for Gemini 3.5 Flash, and 63.9% on MLE Bench against 49.7%. Generated code arrives with fewer unwanted edits and less execution looping, and web and app development output comes back cleaner, which matters when an agent writes a multi-element layout you intend to ship rather than rewrite.

Computer use ships as a built-in client-side tool. Gemini 3.6 Flash scores 83.0% on OSWorld-Verified against 78.4% for Gemini 3.5 Flash, so agents that drive a browser or a desktop application close the gap between a plan and a completed action more often. Multimodal and spatial reasoning improved alongside it, covering chart interpretation, visual blueprint conversion, and multi-element web layout generation. Gemini 3.6 Flash accepts text, images, audio, video, and PDF documents within a context window of 1M tokens and returns text, up to 64K tokens.

Knowledge work moved with the rest: Gemini 3.6 Flash scores 1421 on GDPval-AA v2 against 1349 for Gemini 3.5 Flash, and Google points to document parsing, chart and data analysis, and report drafting as the tasks customers exercise most. Function calling, structured outputs, code execution, search grounding, URL context, and context caching are all supported. Calling Gemini 3.6 Flash through AI Gateway adds usage and cost tracking, automatic retries, and provider failover on one API surface, with no Google Cloud account required.

## What to consider

Gemini 3.6 Flash defaults to the `medium` thinking level. Set `thinkingLevel` under `providerOptions.google.thinkingConfig` to `minimal` or `low` for extraction, routing, and classification, and to `high` when an agent has to reason across many steps. Thinking tokens count toward output tokens, so the level you pick moves the cost of every request. Budget against realistic thinking settings rather than the default before you size a production deployment.

## When to use

### Best For

- **Agentic coding loops** where fewer reasoning steps and tool calls per task cut the cost of a full run
- **Production code generation** with fewer unwanted edits and less execution looping than Gemini 3.5 Flash
- **Computer use agents** that drive browsers and desktop applications through a built-in tool
- **Web and app development** where generated layouts and components need less rewriting
- **Multimodal knowledge work** including chart interpretation, document parsing, and report drafting

### Consider Alternatives When

- The task needs the deepest reasoning available. `google/gemini-3.1-pro-preview` carries more depth on hard multi-step problems
- Per-token cost sets your volume ceiling. `google/gemini-3.5-flash-lite` runs high-throughput extraction and classification more cheaply
- You need image output. `google/gemini-3.1-flash-image` and `google/gemini-3-pro-image` return images; Gemini 3.6 Flash returns text
- You need video output. `google/veo-3.1-generate-001` and `google/gemini-omni-flash-preview` generate video

## Best for

- **Agentic Coding Loops:** Fewer reasoning steps and tool calls per task cut the cost of a full multi-step run
- **Production Code Generation:** DeepSWE and MLE Bench results improve over Gemini 3.5 Flash with fewer unwanted edits
- **Computer Use Agents:** A built-in computer use tool and higher OSWorld-Verified results support browser and desktop automation
- **Web and App Development:** Cleaner layout and component output leaves less generated frontend code to rewrite
- **Multimodal Knowledge Work:** Chart interpretation, document parsing, and report drafting all improved in this release

## Consider alternatives

- **Deepest Reasoning Required:** `google/gemini-3.1-pro-preview` carries more depth on hard multi-step problems
- **Volume Over Capability:** `google/gemini-3.5-flash-lite` runs high-throughput extraction and classification more cheaply
- **Native Image Output:** `google/gemini-3.1-flash-image` and `google/gemini-3-pro-image` return images; Gemini 3.6 Flash returns text
- **Video Generation Work:** `google/veo-3.1-generate-001` and `google/gemini-omni-flash-preview` produce video output

## Frequently asked questions

### What changed in Gemini 3.6 Flash compared to Gemini 3.5 Flash?

Gemini 3.6 Flash improves coding, knowledge work, and multimodal performance while consuming fewer output tokens. It also completes multi-step workflows with fewer reasoning steps and fewer tool calls, and computer use now ships as a built-in tool.

### How do I control how much Gemini 3.6 Flash thinks before it answers?

Set `thinkingLevel` under `providerOptions.google.thinkingConfig`. Gemini 3.6 Flash defaults to `medium`. Use `minimal` or `low` for extraction, routing, and classification, and `high` for autonomous subagents and long tool-use chains.

### Which benchmark results improved over Gemini 3.5 Flash?

Gemini 3.6 Flash scores 49% on DeepSWE against 37%, 63.9% on MLE Bench against 49.7%, 83.0% on OSWorld-Verified against 78.4%, and 1421 on GDPval-AA v2 against 1349. Validate against your own workload before you migrate production traffic.

### Which input types does Gemini 3.6 Flash accept?

Text, images, audio, video, and PDF documents, within a context window of 1M tokens. Output is text only, up to 64K tokens. Gemini 3.6 Flash does not generate images or audio.

### Does Gemini 3.6 Flash support computer use?

Yes, as a built-in client-side tool rather than something you assemble yourself. OSWorld-Verified results improved over Gemini 3.5 Flash, which matters for agents that click through browsers and desktop applications.

### Does Gemini 3.6 Flash support function calling and structured outputs?

Yes. Function calling, structured outputs, code execution, search grounding, URL context, and context caching are all supported. Reach them through the AI SDK, the OpenAI-compatible Chat Completions endpoint, the Responses API, or another supported interface.

### How much does Gemini 3.6 Flash cost on AI Gateway?

Current rates are $0.75 per million input tokens and $3.75 per million output tokens, listed on this page and subject to provider updates. AI Gateway reflects provider pricing with no markup and charges no platform fee on inference.

### How do I call Gemini 3.6 Flash on AI Gateway?

Set the model to `google/gemini-3.6-flash` using the AI SDK, the Chat Completions endpoint, the Responses API, or another supported interface. AI Gateway handles provider routing, retries, and failover, and you do not need a Google Cloud account.

### When should I use Gemini 3.6 Flash instead of Gemini 3.5 Flash-Lite?

Choose Gemini 3.6 Flash when a task needs reasoning depth, code quality, or reliable computer use. Choose `google/gemini-3.5-flash-lite` for high-volume extraction, classification, and scoped subagent steps where per-token cost caps how much you can process.

### How does Zero Data Retention work with Gemini 3.6 Flash through AI Gateway?

Yes, Zero Data Retention is available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.

## Links

- [Model page](https://vercel.com/ai-gateway/models/gemini-3.6-flash)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/google-models)