New: Gemini 3.6 Flash (gemini-3.6-flash) and Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite) are generally available (GA), with dedicated model pages and a Latest models guide (Interactions and generateContent versions) covering capabiliti…
Changelog
Gemini API docs updates
Models
New: Gemini 3.6 Flash (gemini-3.6-flash) and Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite) are generally available (GA), with dedicated model pages and a Latest models guide (Interactions and generateContent versions) covering capabilities, API changes, and migration.
Pricing & specs: 3.6 Flash at $1.50/1M input and $7.50/1M output tokens; 3.5 Flash-Lite at $0.30/1M input and $2.50/1M output. Both support a 1M-token context window, 64k max output, thinking, and built-in tools (Computer Use on 3.6 Flash in preview).
Docs defaults: Examples and highlights across the docs now use gemini-3.6-flash as the primary Flash model; 3.5 Flash-Lite replaces 3.1 Flash-Lite in product positioning.
Agents
Antigravity agent: Default model is now Gemini 3.6 Flash; you can select the underlying model (including 3.6 Flash, 3.5 Flash, or 3.5 Flash-Lite) with agent_config.
API behavior
Deprecated: Sampling parameters temperature, top_p, and top_k are deprecated; see the latest models guide for details.