Gemini 3.5 Flash-Lite

Claims and evidence

• Released 21 July 2026 ● Primary source[1]3.6 Flash and 3.5 Flash-Lite: Get started today [4]July 21, 2026 [5]Versions

• Status Available ● Primary source[6]Gemini 3 models: gemini-3.5-flash-lite [7]Models available for at least 12 months after release

• Successor of Gemini 3.1 Flash-LiteThe model card says 3.5 Flash-Lite is based on Gemini 3.1 Flash-Lite, and the Gemini API deprecations page names gemini-3.5-flash-lite as the recommended replacement for gemini-3.1-flash-lite. ● Primary source[2]Model Information: Model dependencies [6]Gemini 3 models: gemini-3.1-flash-lite [1]3.5 Flash-Lite: Built to scale agentic workflows

• Change · Tool use Computer use is available as a built-in tool; the Gemini API model page lists it as a preview capability.Compared with Gemini 3.1 Flash-Lite ● Primary source[1] [3]Capabilities: Computer use
• Change · Other Google reports higher quality than 3.1 Flash-Lite across thinking levels in coding, agentic, long-context and real-world task evaluations.Compared with Gemini 3.1 Flash-Lite ● Primary source[1]
• Change · Other List price rose to 0.30 USD per million input tokens and 2.50 USD per million output tokens, from 0.25 USD and 1.50 USD for 3.1 Flash-Lite.Compared with Gemini 3.1 Flash-Lite ● Primary source[2]

• Input text, image, audio, videoPDF input is also listed. ● Primary source[3]Supported data types: Inputs [2]Model Information: Inputs
• Output text ● Primary source[3]Supported data types: Output [2]Model Information: Outputs
• Feature Reasoning modeConfigurable thinking levels; Agent Platform says the model defaults to minimal thinking. ● Primary source[3]Capabilities: Thinking [5]introduction: thinking levels
• Feature Computer useListed as a preview capability on the Gemini API model page. ● Primary source[1]3.5 Flash-Lite: Built to scale agentic workflows [3]Capabilities: Computer use
• Context window 1,048,576 tokensThe model card gives a context window of up to 1M tokens; output is limited to 65,536 tokens. ● Primary source[3]Token limits: Input token limit [5]Token limits: Context window
• Parameters Not disclosedNot disclosed by Google. No claim madeNo source
• Access API, consumer appGemini API (Google AI Studio, Android Studio), Gemini Enterprise Agent Platform and the Gemini app; the announcement says it was also rolling out in Google Search. ● Primary source[1]3.6 Flash and 3.5 Flash-Lite: Get started today [2]Distribution

Lineage

Variants

No variants recorded in this record.

Related AI Radar coverage

• Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash CyberGoogle (AI & DeepMind) · DeepMind · 21 July 2026

All model releases from Google on AI Radar →