Google DeepMind · 2026-07-21 · major
Gemini 3.6 Flash — Google's workhorse plus Flash-Lite and Flash Cyber
Gemini 3.6 Flash hits 49% on DeepSWE (up from 37%) and 83% on OSWorld while cutting output tokens 17%. Google also ships Flash-Lite for high-throughput agents and Flash Cyber for code security.
Google ships three Flash-tier Gemini variants tuned for coding, high-throughput agents, and cybersecurity.
Key specs
| Deep swe | 49% |
|---|---|
| Osworld verified | 83.0% |
| Output tokens | -17% |
Quick facts
| Maker | Google DeepMind |
|---|---|
| Models | Gemini 3.6 Flash, 3.5 Flash-Lite, 3.5 Flash Cyber |
| Price (3.6 Flash) | $1.50 in / $7.50 out per 1M tokens |
| Price (3.5 Flash-Lite) | $0.30 in / $2.50 out per 1M tokens |
| Speed (Flash-Lite) | 350 output tokens/sec |
| Availability | Gemini API, AI Studio, Antigravity, Gemini Enterprise, Gemini app |
| Flash Cyber access | Governments and trusted partners via CodeMender |
Benchmarks
| Gemini 3.5 Flash Cyber | 55 issues | |
|---|---|---|
| Gemini 3.5 Flash | 47 issues | |
| Claude Opus 4.6 | 36 issues |
Pricing
| 3.6 Flash input | $1.50 / 1M tokens |
|---|---|
| 3.6 Flash output | $7.50 / 1M tokens |
| 3.5 Flash-Lite input | $0.30 / 1M tokens |
| 3.5 Flash-Lite output | $2.50 / 1M tokens |
What is it?
Gemini 3.6 Flash is Google's new workhorse Flash model, aimed at coding, knowledge work, and multimodal tasks. Alongside it Google released Gemini 3.5 Flash-Lite, a low-latency variant hitting 350 output tokens per second, and Gemini 3.5 Flash Cyber, a security-tuned model limited to CodeMender and trusted partners.
How does it work?
The 3.6 Flash update trims output token usage by 17% while lifting scores across agent and coding evals. Flash-Lite trades a bit of quality for throughput on high-volume agent workflows. Flash Cyber is fine-tuned on cybersecurity data so it can scan large codebases in many parallel passes, finding 55 confirmed V8 issues on the Big Sleep evaluation versus 47 for stock 3.5 Flash and 36 for Claude Opus 4.6.
Why does it matter?
Google is filling the middle of its lineup with cheaper, faster models that still handle real agent work. Flash 3.6 costs $1.50 input and $7.50 output per 1M tokens, while Flash-Lite drops to $0.30 in and $2.50 out. That pricing plus the DeepSWE, MLE Bench, and OSWorld gains makes Flash a viable default for production coding agents that used to need a larger tier.
Who is it for?
Developers and enterprises building production AI agents
Frequently asked questions
- How much does Gemini 3.6 Flash cost?
- Gemini 3.6 Flash is priced at $1.50 per 1M input tokens and $7.50 per 1M output tokens on the Gemini API. The lower Gemini 3.5 Flash-Lite tier costs $0.30 input and $2.50 output per 1M tokens for teams running high-volume agent workflows.
- Where can I use Gemini 3.6 Flash?
- Gemini 3.6 Flash is available through the Gemini API in Google AI Studio and Android Studio, Google Antigravity, the Gemini Enterprise Agent Platform, and the consumer Gemini app. Flash-Lite is rolling out to the same API plus Google Search.
- How does Gemini 3.6 Flash compare to Gemini 3.5 Flash?
- Gemini 3.6 Flash jumps from 37% to 49% on DeepSWE, from 49.7% to 63.9% on MLE Bench, and from 78.4% to 83.0% on OSWorld-Verified, while cutting output tokens by 17%. Google frames it as a workhorse replacement for the older 3.5 Flash.
- What is Gemini 3.5 Flash Cyber for?
- Gemini 3.5 Flash Cyber is a security-tuned model built to discover, validate, and patch software vulnerabilities inside Google's CodeMender agent. On Big Sleep it flagged 55 confirmed V8 issues versus 47 for stock 3.5 Flash and 36 for Claude Opus 4.6. Access is limited to governments and trusted partners.
- Can I get the weights for Gemini 3.6 Flash?
- No. Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber are closed-weight Google models offered only through the Gemini API and Google's own products. Flash Cyber has an even tighter access list, gated to governments and trusted partners via CodeMender.
Try it
https://ai.google.dev/gemini-api/docs/models