AI Model Release Tracker

Gemini 3.5 Flash-Lite

Released

Gemini 3.5 Flash-Lite is an AI model released by Google on Tuesday, Jul 21 2026, the same day as Gemini 3.6 Flash. It has a 1M token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover BullshitBench v2, SWE-Bench Pro, Terminal-Bench 2.1, BU Bench, OSWorld-Verified, and GDPval-AA v2.

API pricing

InputWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it.
$0.30
Cached inputA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again.
$0.03
OutputWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one.
$2.50
  • Output prices include reasoning tokens.
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Verified October 19, 2026Read from ai.google.dev. Click the check to open the source.

Available from

Googleglobalflex$0.15$1.251M
Google AI Studio—flex$0.15$1.251M
Googleglobal—$0.30$2.501M
Google AI Studio——$0.30$2.501M
Googleeu—$0.33$2.751M
Googleus—$0.33$2.751M
Googleglobalpriority$0.54$4.501M
Google AI Studio—priority$0.54$4.501M
USD per 1M tokensEvery price here is for one million tokens. A token is roughly three-quarters of a word, so a million tokens is about 750,000 words of text.Checked October 20, 2026Third-party rates from openrouter.ai, not the lab’s own list price.

Benchmarks

Coding

SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better.
54.2%
#17 of 24

Terminal & CLI

Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better.
54%
#34 of 36

Agentic & tool use

BU BenchBrowser agent — Can the AI drive a real web browser to finish tasks — clicking, filling forms, and navigating sites the way a person would? Run by Browser Use on their BU Bench task set. Higher is better.
49%
#6 of 7
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
74%
#14 of 29

Knowledge work

GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo)
1140
#19 of 21

Robustness

BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better.
66%
#19 of 97

Compare Gemini 3.5 Flash-Lite with

Gemini 3.5 Flash-Lite

Suggested comparisons

Gemini 3.5 Flash-LitevsGemini 4 ArgonGemini 3.5 Flash-LitevsGPT-6.1 SolGemini 3.5 Flash-LitevsClaude Haiku 5.5Gemini 3.5 Flash-LitevsMuse Spark OpenGemini 3.5 Flash-LitevsGrok 4.7Gemini 3.5 Flash-LitevsDeepSeek-V4.1-Flash-1017Gemini 3.5 Flash-LitevsMistral Large 4Gemini 3.5 Flash-LitevsKimi K3Gemini 3.5 Flash-LitevsGLM-5.3-FlashGemini 3.5 Flash-LitevsQwen3.8-Max-0902Gemini 3.5 Flash-LitevsNemotron 3.5 Super VL

About

Gemini 3.5 Flash-Lite joined Google's model lineup in July 2026 as an efficiency-focused model for high-throughput tasks. Google positioned it for document parsing, lightweight agent work and frequent data-processing requests. It accepted text and multimodal inputs, including images, audio, video and PDFs, while producing text responses rather than generating media.

Its documented input limit was roughly one million tokens, with support for functions and structured output. That combination suited workflows such as extracting fields from files or assigning routine steps to a smaller model within a larger agent system. The Flash-Lite tier offered a different deployment tradeoff from the main Flash model, with task complexity, response time and per-request cost guiding the choice.

Frequently asked questions

Gemini 3.5 Flash-Lite was released by Google on Tuesday, Jul 21 2026.

All Google releases

32 tracked