Developers and agents building AI clients need higher token efficiency, lower latency, and more reliable performance. Our Flash series of models are designed to meet the sweet spot of efficiency and quality to enable agent workflow. Building on Gemini 3.5 Flash, we introduce new Gemini models:
- 3.6 Flash: Our workhorse model that delivers better coding, knowledge work, and multimodal performance. According to the Artificial Analysis Index, it reduces the consumption of output tokens by 17% compared to 3.5 Flash, and in some benchmarks like DeepSWE with Datacurve, we see up to 65%, all at a lower cost per output token.
- 3.5 Flash-Lite: Our fastest, cost-effective 3.5 class model delivers 350 output tokens per second according to the Artificial Analysis Index, and outperforms previous generations of Flash-Lite in agent workflow.
- 3.5 Flash Cyber in CodeMender: Successful cybersecurity applications require careful modeling around the agent infrastructure. Introducing the combination of a new, high-performance, specialized Internet-centric model paired with our CodeMender code security agent that delivers competitive edge performance.
Beyond today's release, Gemini 3.5 Pro is currently in testing with our partners and we plan to make it widely available as soon as it's ready. In parallel, our team is already focused on building the next generation of models. We've started our much-desired early training run, on Gemini 4, and we're excited about the progress.
3.6 Flash: Works well and better quality than 3.5 Flash
Gemini 3.6 Flash builds directly on developer and customer feedback from 3.5 Flash. 3.6 Flash not only brings a step forward in coding and information, but it does this while improving the efficiency of tokens. For example, in the Artificial Analysis Index, we see 3.6 Flash consumes 17% less output tokens than 3.5 Flash. It also takes fewer thought steps and tool calls to accomplish a multi-step workflow.
This improved performance is also combined with a lower price than 3.5 Flash. At $1.50/1M input tokens and $7.50/1M output tokens, 3.6 Flash reduces the overall cost per agent transaction, making agents less expensive to build and operate.
