Models·Google

Gemini 3.5 Flash Lite

google/gemini-3.5-flash-lite

Google's efficient model for focused tasks in complex, multi-agent workflows.

$0.30Input, per million tokens
$2.50Output, per million tokens
1.05MContext window
66KLongest answer

1.05M tokens of context — about 786,432 words, on a scale that ends at 2M.

What it can do

  • Reasoning
  • Vision
  • Tool use
  • Takes text
  • Takes image
  • Takes video
  • Takes file
  • Takes audio
  • Gives text

Released 21 Jul 2026 · checked 3 Oct 2026.

About it

Gemini 3.5 Flash Lite is a Google model for focused work inside complex, multi-agent workflows. It supports reasoning, tool use, and vision, and accepts text, images, video, files, and audio. Its large context window can read up to 1,048,576 tokens, while answers can reach 65,536 tokens.

Review

In a line

Gemini 3.5 Flash Lite is a Google model designed for focused tasks in larger, multi-agent workflows. It produces text and supports reasoning, tools, and vision. See Google's other models.

What it's good at

This model suits subagents that handle a defined task while working as part of a complex workflow. Its reasoning ability can help with tasks that require step-by-step decisions.

It can work with text, images, video, files, and audio as input. Tool support and vision add ways to handle tasks beyond plain text. Compare vision language models and reasoning models.

Price in plain terms

Per million tokens, input is priced at $0.3 and output at $2.5, so generated text costs more than text sent to the model.

Reading a roughly 500-word page costs about 0.019 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.15 cents. 1,000 such turns cost about $1.55.

Context and limits

The context window is 1,048,576 tokens. That is the amount of material the model can consider in one request. It can produce up to 65,536 tokens in one answer.

Its output is text only, even though it accepts text, images, video, files, and audio.

Ways to use it

Use it where an app, chat tool, or provider makes Gemini 3.5 Flash Lite available. The model ID is google/gemini-3.5-flash-lite.

Its inputs can include text, images, video, files, and audio, making it suitable for focused tasks that need more than text alone.

Verdict

Choose this model when a workflow needs a focused worker that can reason, use tools, and inspect visual or other media inputs. The very large context window also fits tasks that involve lengthy material.

Look elsewhere if you need outputs in a format other than text, or if your work does not benefit from multimodal input, reasoning, or tool support.

Verdict

Gemini 3.5 Flash Lite fits teams building workflows where separate agents handle narrowly defined jobs. It can take many kinds of input, use tools, and reason through a task while keeping costs low for typical chat volumes. The model is less suitable for anyone who needs non-text output or a simple text-only model without multimodal handling. Check that your chosen app or provider offers this exact model before planning a workflow around it.

  • Accepts text, images, video, files, and audio
  • Supports reasoning, tools, and vision
  • Context window of 1,048,576 tokens
  • Typical 1,000 chat turns cost about $1.55
  • Produces text output only
  • Output tokens cost more than input tokens
  • Best suited to focused subagent tasks, not every use case
  • Pick it ifFocused subagents in complex, multi-agent workflows; Tasks that combine reasoning, tools, and visual inputs; Long documents or other large input sets
  • Skip it ifYou need image, audio, or video output rather than text; Your work does not use multimodal inputs or tool calls; Your provider or app does not offer this exact model

Pricing

Input is $0.3 per million tokens and output is $2.5 per million tokens. A typical 1,000-turn example costs about $1.55, with output priced higher than input.

Questions

Who makes Gemini 3.5 Flash Lite?

Google makes it. Its model ID is google/gemini-3.5-flash-lite.

What can Gemini 3.5 Flash Lite read?

It accepts text, images, video, files, and audio as inputs.

Can it use tools?

Yes. The facts list tool support and reasoning.

How much text can it consider at once?

Its context window is 1,048,576 tokens.

How long can one answer be?

The maximum output is 65,536 tokens, and the output format is text.

What does it cost?

Input costs $0.3 per million tokens; output costs $2.5 per million tokens. About 1,000 example chat turns cost $1.55.

On these boards

More from Google