AI Radar

Gemini 3.5 Flash

Available · Language, Multimodal, Reasoning

Gemini 3.5 Flash is a multimodal language model in Google's Gemini Flash line and the first model of the Gemini 3.5 series, generally available from 19 May 2026 in the Gemini API, Agent Platform, the Gemini app and AI Mode in Search, where it became the default. Google positions it for agentic, coding and long-running tasks, including running subagents. [1] Primary source

Timeline of Gemini 3.5 Flash →

Claims and evidence

  • Released Generally available from the first day, at Google I/O 2026. Primary source[1] [2]May 19, 2026 [3]gemini-3.5-flash: release date [4]gemini-3.5-flash: release date
  • Status AvailableStable model with no shutdown date announced in the Gemini API; Agent Platform lists retirement on May 19, 2027 or later. The models overview calls it Google's legacy Flash model on 2026-10-01. Primary source[5]Gemini 3, Stable [3]Gemini 3 models: gemini-3.5-flash [4]Models available for at least 12 months after release
  • Successor of Gemini 3 FlashThe model card says Gemini 3.5 Flash is based on Gemini 3 Flash; the developer guide covers migration from Gemini 3 Flash Preview; gemini-flash-latest moved from gemini-3-flash-preview to this model. Primary source[6]Model Information: Model dependencies [7]Migrate from Gemini 3 Flash Preview [2]May 19, 2026
  • Change · Availability Generally available as a stable model from release, whereas Gemini 3 Flash remained a preview model.Compared with Gemini 3 Flash Primary source[7] [5]Gemini 3, Stable; Gemini 3, Preview
  • Change · Reasoning The default thinking effort is medium instead of high.Compared with Gemini 3 Flash Primary source[7]Behavioral changes
  • Change · Reasoning Keeps intermediate reasoning across multi-turn conversations automatically.Compared with Gemini 3 Flash Primary source[7]
  • Change · Safety Google says the Gemini 3.5 series was developed under its Frontier Safety Framework, with strengthened cyber and CBRN safeguards.Compared with Gemini 3 Flash Primary source[1]
  • Input text, image, audio, videoPDF input is also supported. Primary source[8]Supported data types: Inputs [6]Inputs
  • Output text Primary source[8]Supported data types: Output
  • Feature Reasoning modeThinking levels minimal, low, medium (default) and high. Primary source[8]Capabilities: Thinking [7]Behavioral changes
  • Feature Function calling Primary source[8]Capabilities: Function calling
  • Feature Computer useComputer Use tool in public preview from 2026-06-24. Primary source[2]June 24, 2026 [8]Capabilities: Computer use
  • Context window 1,048,576 tokensInput token limit; the output limit is 65,536 tokens. The model card gives a context window of up to 1M. Primary source[8]Token limits: Input token limit
  • Parameters Not disclosedNot disclosed by Google. No claim madeNo source
  • Access API, consumer appAPI: the Gemini API (Google AI Studio, Android Studio), Google Antigravity and Gemini Enterprise Agent Platform. Consumer: the Gemini app and AI Mode in Search, where it became the default model. Primary source[1]3.5 Flash is available today

Lineage

Predecessors

Successors

Based on

Not derived from another model.

Variants and derived

None recorded.

Siblings

None recorded.

All ancestors

All descendants

Variants

No variants recorded in this record.

Related AI Radar coverage

All model releases from Google on AI Radar →