Skip to content

Google · Chat

Gemini 3.1 Flash Lite

Google 🇺🇸ChatGemini familyOne of its sources says the weights aren't publicly available.The maker currently plans to deprecate it within the next 180 days (7 May 2027). Plans can change.

Newer version: Gemini 3.5 Flash Lite. We compared release dates and version names in our catalogue.

From Neosantara's listing:
Google Gemini 3.1 Flash Lite model for fast and efficient tasks.
llminfo IQ
125.3
Max context
1M
Listings
63

Evidence and sources

llminfo IQ 125.3, ranked among chat models. From 275 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 97.2, 95% interval 88–107. From 16 benchmark families across 8 sources. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedGood value (55.5). Cheaper per task than its benchmark results predict. From 32 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksYap score: tokens used per coding task vs. models with the same resultsTypical (48 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. By effort: default 48 · minimal 43 · low 44 · medium 41 · high 45.Median 16,242 output tokens per task at default effort.How the Yap score worksReleased 3 Mar 2026Maker HQ: US (according to Google's own website)
All rankings: #41 on OpenRouter this month #50 on OpenRouter today #50 on OpenRouter this week #93 on OpenRouter today #98 on OpenRouter this week #100 on OpenRouter this month #118 on NanoGPT recentlySources: Google Generative AI API data docs, DeepInfra data, LiteLLM data, Models.dev data, AI/ML API data, Neosantara Inference data, llminfo researchPrimary data from Google Generative AI API from DeepInfra: variant from LiteLLM: deprecation date from Models.dev: weights availability from AI/ML API: primary mode from Neosantara Inference: description from llminfo research: release date
Tags from its sources:instructtoolsvisionreasoningaudiodocument

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Trade-offs

Deprecation scheduled

The maker plans to deprecate it on 7 May 2027, within the next 180 days.

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.

Input & output

as stated by the model's providers
Text
Images
Audio
Video
Files
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingPDF inputWeb searchAudio inputBatch API

Listings

 Googlemodels/gemini-3.1-flash-liteSources: GenAI Prices data, Google Generative AI API data docs, Google Vertex AI Pricing, LiteLLM data, Models.dev data1M$0.25>200k $0.25cache read $0.025batch $0.125$1.50>200k $1.50batch $0.75$0.50$0.05—
 Nous Portalgoogle/gemini-3.1-flash-lite:batch · batchSource: Nous Portal Inference Gateway data1M$0.125cache read $0.0125$0.75$0.25—$0.00000013
 OpenRouterInference by Googlegoogle/gemini-3.1-flash-lite:batch · batch, globalSource: OpenRouter data docs1M$0.125cache read $0.0125$0.75$0.25$0.025$0.00000013
 Vercelgoogle/gemini-3.1-flash-lite-batch · batchSource: Vercel AI Gateway data1M$0.125cache read $0.015$0.75———
 Meganovagemini/gemini-3.1-flash-liteSource: MegaNova data1M$0.20$1.20———
 Tokenmixgemini-3.1-flash-liteSources: TokenMix data, TokenMix PricingNot published by the provider$0.2425$1.455———
 302aigemini-3.1-flash-lite-previewSources: 302.ai data, Models.dev data1M$0.25cache read $0.03$1.50———
 Aihubmixgemini-3.1-flash-liteSource: AIHubMix data docs1M$0.25cache read $0.025$1.50———
 Atlascloudgoogle/gemini-3.1-flash-liteSource: AtlasCloud data1M$0.25cache read $0.025$1.50———
 Cometapigemini-3.1-flash-liteSources: CometAPI data, CometAPI Pricing dataNot published by the provider$0.25$1.50———

8 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Human preference

Arena ELO63 related benchmarks · 63 results
Source: LMArena data · Nov 2026
BenchmarkScore
Arena ELO
Chinese
Creative Writing
English
Show 59 more benchmarksHide 59 benchmarks
BenchmarkScore
Exclude Ties
Expert
French
German
Hard Prompts
Hard Prompts English
Industry Business and Management and Financial Operations
Industry Entertainment and Sports and Media
Industry Life and Physical and Social Science
Industry Mathematical
Industry Medicine and Healthcare
Industry Software and It Services
Industry Writing and Literature and Language
Japanese
Korean
Longer Query
Multi Turn
Non English
Polish
Russian
Spanish
Style-Controlled
Text Factuality
Text Factuality - Chinese
Text Factuality - Creative Writing
Text Factuality - English
Text Factuality - Exclude Ties
Text Factuality - Expert
Text Factuality - French
Text Factuality - German
Text Factuality - Hard Prompts
Text Factuality - Hard Prompts English
Text Factuality - Industry Business and Management and Financial Operations
Text Factuality - Industry Entertainment and Sports and Media
Text Factuality - Industry Life and Physical and Social Science
Text Factuality - Industry Mathematical
Text Factuality - Industry Medicine and Healthcare
Text Factuality - Industry Software and It Services
Text Factuality - Industry Writing and Literature and Language
Text Factuality - Japanese
Text Factuality - Korean
Text Factuality - Longer Query
Text Factuality - Multi Turn
Text Factuality - Non English
Text Factuality - Polish
Text Factuality - Russian
Text Factuality - Spanish
Vision
Vision - Captioning
Vision - Chinese
Vision - Creative Writing Vision
Vision - Diagram
Vision - English
Vision - Entity Recognition
Vision - Homework
Vision - Humor
Vision Style-Controlled
Design Arena EloSource: Design Arena Leaderboards data · Nov 2026

Long context

OpenAI MRCR v2 · 8-needleGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · 128K cumulative average · Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Coding

Arena ELOSource: LMArena data · Nov 2026
LiveCodeBenchGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · 175 problems · 2025-01-01 through 2025-05-01 · Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
ValsSource: Vals AI

Reasoning

Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · no tools
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Agentic

Source: OpenRouter benchmark exports · Oct 2026
7 May 2026 version · 90-day results
Across 2 providers lowest 69% median 70.8% highest 72.7%OpenRouter's automatic routing 71.3%
Show results from 2 providers and OpenRouter's automatic routingHide results from 2 providers and OpenRouter's automatic routing
  • Google
  • OpenRouter's automatic routing
  • Google AI Studio

Science

GPQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
GPQA Diamond7 results
Source: Epoch AI Benchmarking Hub data
Reasoning effort
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · no tools
Source: OpenRouter benchmark exports · Oct 2026
7 May 2026 version · 90-day results
Across 2 providers lowest 81.3% median 81.4% highest 81.4%OpenRouter's automatic routing 82.1%
Show results from 2 providers and OpenRouter's automatic routingHide results from 2 providers and OpenRouter's automatic routing
  • OpenRouter's automatic routing
  • Google AI Studio
  • Google

General

Math

Vision

CharXiv ReasoningGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · reasoning · no tools · Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
CharXiv-RSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MMMU-Pro2 results
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · no tools
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
VideoMMMU2 results
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Factuality

FACTS Benchmark SuiteGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SimpleQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
SimpleQA VerifiedGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Instruction following

Safety

PhareSource: Phare

Multilingual

MMMLU2 results
Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Gemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Knowledge

OCR

Other

Horangi36 related benchmarks · 36 results
MRCR v2 · 1M PointwiseGemini 3.1 Flash-Lite Model Card · high thinking · default sampling · single attempt/pass@1 · 1M pointwise · Reported by Google DeepMind · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Nejumi 428 related benchmarks · 28 results
ALT_バイアス
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
ALT_倫理・道徳
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
ALT_制御性
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
ALT_堅牢性
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
ALT_毒性
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
ALT_真実性
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
Show 22 more resultsHide 22 results
AVG_jaster_0shot
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
AVG_jaster_2shots
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_アプリケーション開発
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_コーディング
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_一般的知識
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_基礎的言語性能
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_専門的知識
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_応用的言語性能
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_情報検索
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_意味解析
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_抽象的推論
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_推論能力
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_数学的推論
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_構文解析
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_知識・質問応答
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_翻訳
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_表現
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_論理的推論
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
GLP_関数呼び出し
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
TOTAL_SCORE
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
アラインメント(ALT)_AVG
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
汎用的言語性能(GLP)_AVG
Source: Nejumi Leaderboard 4 data docs
google/gemini-3.1-flash-lite-preview
Vals10 related benchmarks · 12 results
EMB
Source: Vals AI
CorpFin v2
Source: Vals AI
LegalBench
Source: Vals AI
MedCode
Source: Clinical Benchmarks data docs · Sep 2026
Setting not stated
Source: Vals AI
Setting not stated
Show 5 more resultsHide 5 results
MedScribe
Source: Clinical Benchmarks data docs · Sep 2026
Setting not stated
Source: Vals AI
Setting not stated

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · LMArena, CC BY 4.0 (licence) · BullshitBench by Peter Gostev, MIT (licence) · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats · Steel.dev Agent Leaderboards, MIT (licence) · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results · EQ-Bench (Samuel J. Paech) — cite paech2023eqbench · LLM Chess by Maxim Saplin, Apache-2.0 (Copyright [2024] [Maxim Saplin]) (licence) · MineBench, MIT (licence) · Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results · Phare (Giskard), CC BY 4.0 (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · Vectara Hallucination Leaderboard, Apache-2.0 (licence)

Around the web

Also known as: databricks-gemini-3-1-flash-lite, databricks/databricks-gemini-3-1-flash-lite, deepinfra/google/gemini-3.1-flash-lite, gemini-3-1-flash-lite, gemini-3.1-flash-lite-nothink, gemini-3.1-flash-lite-preview, gemini-3.1-flash-lite-preview-thinking, gemini-3.1-flash-lite-thinking, gemini-3.1-flash-lite@eu, gemini/gemini-3.1-flash-lite, gemini/gemini-3.1-flash-lite-preview, gmi/google/gemini-3.1-flash-lite-preview, google-ai-studio/gemini-3.1-flash-lite, google-gemini-3.1-flash-lite, google-vertex/gemini-3.1-flash-lite, google/gemini-3-1-flash-lite, google/gemini-3-1-flash-lite-preview, google/gemini-3.1-flash-lite, google/gemini-3.1-flash-lite-batch, google/gemini-3.1-flash-lite-preview, google/gemini-3.1-flash-lite:batch, perplexity/google/gemini-3.1-flash-lite, vertex/gemini-3.1-flash-lite, vertex/gemini-3.1-flash-lite@eu, vertex/gemini-3.1-flash-lite@us, vertex_ai/gemini-3.1-flash-lite, vertex_ai/gemini-3.1-flash-lite-preview

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works