Gemini 3.1 Flash Lite

Gemini 3.1 Flash-Lite is positioned by Google as the fastest and most cost-efficient entry in its Gemini 3 series, with general availability announced on May 7, 2026 through a Google Cloud blog post signed by the VP of Product Management for Gemini Enterprise. The model is designed around three priorities: ultra-low latency, the ability to handle high-volume tasks, and strong cost efficiency for production deployments that need to scale. It sits alongside the broader Pro and Flash lineup as the lightweight option when raw intelligence is less critical than responsiveness and throughput.

Google describes the model as well suited to agentic and software development workflows, where teams need real-time responsiveness for complex code completion, seamless UX design, and agentic developer tools. Developers and enterprises cited by Google have found Flash-Lite precise enough for tool calling and orchestration while remaining affordable enough to run automated pipelines at scale. A separate image-focused sibling, Gemini 3.1 Flash-Lite Image, also documented under the Gemini Enterprise Agent Platform, expands the family into multimodal generation use cases alongside the base Flash-Lite variant.

Impossiblgoogle/gemini-3.1-flash-litegemini-flash-lite

Open provider page Provider docs

Quick Info

Provider: Impossibl
Model key: google/gemini-3.1-flash-lite
Release date: May 7, 2026
Last updated: May 7, 2026
Knowledge cutoff: 2025-01
Input modalities
Output modalities
Capabilities

Cost

Input token cost: $0.25
Output token cost: $1.50

Limits

Output tokens: 65,536 tokens
Context window: 1,048,576 tokens

Transparent token rates

Compare Gemini 3.1 Flash Lite pricing

Rates are shown per one million tokens. Combined means one million input plus one million output tokens.

Browse this family

Latest news about Gemini 3.1 Flash Lite

DevPass (LLM Gateway)

Coverage

This article by hidekazu-konishi.com compiles a comprehensive Google Gemini model release timeline, tracing the family from earlier language models such as LaMDA and PaLM, through the launch of Gemini 1.0 in December 2023, and continuing through the Gemini 1.5, 2.0, 2.5, and Gemini 3 generations, including the Gemini 3 Within the timeline context, Gemini 3.1 Flash-Lite fits as part of the Gemini 3 generation, which the article describes as introducing Fast-Lite variants alongside the Pro and Flash releases, with the broader Gemini family having matured to include adjustable reasoning depth and standardized developer controls. The art

Abacus

CoverageBenchmark

Lumina's capabilities index, snapshot dated September 20, 2026, explicitly names Gemini 3.1 Flash-Lite across four capability areas using 12 retained native records drawn from 11 evidence families. The model scores 104.6 index points in mathematical and scientific problem solving (90% research range 100.04-109.10) and Two of the four tracked capability categories, software engineering and factual reliability, currently have no qualified category score, and Lumina explicitly notes that Gemini 3.1 Flash-Lite lacks an exact base-profile release with a documented 10K output measurement, so other effort, protocol, or alias records are no

DevPass (LLM Gateway)

Coverage

Google announced that Gemini 3.1 Flash-Lite reached general availability on May 7, 2026 via the Google Cloud Blog, authored by VP of Product Management Michael Gerstenhaber. The post explicitly names the "Gemini 3.1 Flash-Lite" variant and positions it as the fastest and most cost-efficient model in the Gemini 3 series The GA announcement includes two enterprise adoption examples that illustrate intended workloads. JetBrains' Director of AI Vladislav Tankov is quoted saying that integrating Gemini 3.1 Flash-Lite transformed the responsiveness of the IDE AI assistant and Junie agent, citing the balance of high intelligence and minimal

Impossibl

Coverage

OpenRouter lists Gemini 3.1 Flash Lite as Google's GA high-efficiency multimodal model, released on May 7, 2026 after its March 3 preview. It supports text, image, video, audio, and PDF inputs within a 1M token context window, and offers full controllable thinking levels (minimal, low, medium, high) so developers can t The page surfaces a per-provider price matrix across Google AI Studio (Flex, standard, Priority) and Vertex variants, with listed input prices ranging from $0.125 to $0.45 per 1M tokens and output prices from $0.75 to $2.70. Effective weighted-average prices shown are $0.1713/M input and $1.49/M output, reflecting cach