Skip to main content
AI Choice Engine

Google

Gemini 3 Flash (Preview)

Gemini 3 Flash (Preview) is a Google model with 1.05M context, $0.50 in / $3 out per million tokens, last verified October 2, 2026. Open weights: no. Status: Legacy; the current pick is Gemini 3.6 Flash. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

Google’s earlier Gemini 3 Flash preview endpoint for multimodal understanding, coding and long-context tasks.

Recorded profile verification October 2, 2026Individual provider fields may changeAccess: preview — Still listed as Preview. The current deprecation table announces no shutdown date and recommends Gemini 3.6 Flash as the replacement.

AI model specification details
SpecificationGemini 3 Flash (Preview)
ProviderGoogle
TierFrontier
Context window1.05M
Max output66K
Input / 1M tokens$0.50
Output / 1M tokens$3
WeightsClosed
ParametersNot verifiedUnverified
Reasoning levelsminimal, low, medium, high
Modalitiestext, image, video, audio, pdf
API model idgemini-3-flash-preview
ReleasedDecember 17, 2025

Pricing tiers: Standard USD/MTok: input $0.50 text/image/video or $1 audio, output including thinking $3; cache reads $0.05 or $0.10 audio, plus $1/MTok/hour storage. Batch and Flex input $0.25 or $0.50 audio, output $1.50, with Standard cache/storage rates. Priority input $0.90 or $1.80 audio, output $5.40, cache $0.09 or $0.18 audio, storage $1.80/MTok/hour. Search grounding shares 5,000 free monthly requests across Gemini 3.x, then $14/1,000; Maps has separate shared free limits and query fees. Limited Standard free-tier access is not an unlimited paid API entitlement.

Best for

  • Agentic coding
  • Multimodal understanding
  • Long-context tasks

Watch out

Published December 17, 2025; remains Preview with no shutdown date announced. Thinking levels minimal/low/medium/high default to high; minimal does not guarantee thinking is off. The exact card allows text/image/video/audio/PDF input and text output, 1,048,576 input and 65,536 output tokens; no audio or image generation or Live API. Pricing calls it legacy and recommends newer Flash models; do not substitute another Flash SKU as this endpoint.

Common questions

Gemini 3 Flash (Preview)

Answered from the verified figures on this page rather than general guidance.

How much does Gemini 3 Flash (Preview) cost per million tokens?
Gemini 3 Flash (Preview) lists base rates of $0.50 per million input tokens and $3 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. Standard USD/MTok: input $0.50 text/image/video or $1 audio, output including thinking $3; cache reads $0.05 or $0.10 audio, plus $1/MTok/hour storage. Batch and Flex input $0.25 or $0.50 audio, output $1.50, with Standard cache/storage rates. Priority input $0.90 or $1.80 audio, output $5.40, cache $0.09 or $0.18 audio, storage $1.80/MTok/hour. Search grounding shares 5,000 free monthly requests across Gemini 3.x, then $14/1,000; Maps has separate shared free limits and query fees. Limited Standard free-tier access is not an unlimited paid API entitlement.
What is Gemini 3 Flash (Preview)'s context window?
Gemini 3 Flash (Preview) accepts about 1.05M tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.
What is Gemini 3 Flash (Preview) best for?
Gemini 3 Flash (Preview) is a frontier tier from Google. It suits agentic coding, multimodal understanding, long-context tasks. Published December 17, 2025; remains Preview with no shutdown date announced. Thinking levels minimal/low/medium/high default to high; minimal does not guarantee thinking is off. The exact card allows text/image/video/audio/PDF input and text output, 1,048,576 input and 65,536 output tokens; no audio or image generation or Live API. Pricing calls it legacy and recommends newer Flash models; do not substitute another Flash SKU as this endpoint.