← All models

Gemini 3.5 Flash

Published Publisher Googlegemini-3.5-flash

Overview

Gemini 3.5 Flash is a fast, natively multimodal reasoning model in the Gemini 3 series, based on Gemini 3 Flash. Google positions it for coding and complex agentic workflows. Adjustable thinking levels let developers balance quality, cost and latency. (blog.google)

Capabilities and uses

  • Accepts text, images, audio and video; generates text. Google Cloud documents a 1,048,576-token context window and a maximum of 65,536 output tokens. (docs.cloud.google.com)
  • Supports features including structured output, context caching and tool use. Potential uses include software development, document analysis and multi-step enterprise workflows. (docs.cloud.google.com)
  • Google reports a score of 76.2% on Terminal-Bench 2.1 and 83.6% on MCP Atlas in its May 2026 evaluations; these are benchmark results under the stated evaluation conditions, not guarantees of real-world performance. (deepmind.google)

Limitations

The model produces text rather than generated images or audio. Google Cloud lists agentic video understanding as unsupported. Outputs and consequential agent actions should be reviewed; long-context performance can vary by task. (ai.google.dev)

Pricing

The official Gemini Developer API pricing page lists the following USD rates for gemini-3.5-flash, per 1 million tokens. Pricing checked on November 16, 2026; the indexed official pricing page was available, although the live page could not be opened directly during this check. (ai.google.dev)

  • Standard paid tier: input $1.50; output $9.00, including thinking tokens.
  • Batch paid tier: input $0.75; output $4.50, including thinking tokens.
  • Paid context caching: cached input $0.15 per 1 million tokens; storage $1.00 per 1 million tokens per hour.
  • Free tier: standard input and output are listed as free; batch is not available in the free tier. Usage limits and other service charges may apply.

These are Gemini Developer API rates, not necessarily Google Cloud regional or other platform rates. Official pricing source: Gemini Developer API pricing. (ai.google.dev)

Release and official sources