llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model Directory

Scoring & Data

Scoring & Data
1450 models809 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

Google model product

Gemma 4 31B

Gemma 4 31B is a Google dense multimodal model with 31 billion parameters and a 256K context window.

Updated Nov 18, 2026. Default version: Gemma 4 31B

LLMBoard Score57.3Gemma 4 31B
Coverage60%12 benchmark families
Context window262.1KTokens
Official input priceN/AOfficial price unavailable

On this page

  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Runtime
  • Specification
  • Versions
  • Similar models
  • About
  • FAQ

Gemma 4 31B Capability Profile

This profile uses the model's current scored version. Arena ratings and prices are shown separately.

Gemma 4 31B LLMBoard score breakdown

Gemma 4 31B Benchmark Results

Benchmark scores for Gemma 4 31B.

27 rows
Columns

Show columns

Sort by
Benchmark
Score
Rank
Participants
Percentile
Evidence
Evaluated
BenchmarkBIG-Bench Extra HardScore74.40%Rank01Participants11Percentile100.00%EvidenceCEvaluatedNov 17, 2026
BenchmarkMedXpertQAScore61.30%Rank05Participants12Percentile63.64%EvidenceCEvaluatedNov 17, 2026
Benchmarkt2-benchScore86.40%Rank05Participants23Percentile81.82%EvidenceCEvaluatedNov 17, 2026
BenchmarkMRCR v2 (8-needle)Score66.40%Rank08Participants25Percentile70.83%EvidenceCEvaluatedNov 17, 2026
BenchmarkMathVisionScore85.60%Rank13Participants40Percentile69.23%EvidenceCEvaluatedNov 17, 2026
BenchmarkMMMLUScore88.40%Rank18Participants51Percentile66.00%EvidenceCEvaluatedNov 17, 2026
BenchmarkMMLU-ProScore85.20%Rank19Participants142Percentile87.23%EvidenceCEvaluatedNov 17, 2026
BenchmarkAIME 2026Score89.20%Rank20Participants28Percentile29.63%EvidenceCEvaluatedNov 17, 2026
BenchmarkAA IFBenchScore75.58%Rank21Participants168Percentile88.02%EvidenceBEvaluatedN/A
BenchmarkLM Arena Document Style ControlScore1,452.04 ratingRank26Participants39Percentile34.21%EvidenceAEvaluatedNov 9, 2026
BenchmarkMMMU-ProScore76.90%Rank27Participants77Percentile65.79%EvidenceCEvaluatedNov 17, 2026
BenchmarkLM Arena VisionScore1,279.39 ratingRank28Participants113Percentile75.89%EvidenceAEvaluatedNov 9, 2026
BenchmarkLiveCodeBench v6Score80.00%Rank29Participants69Percentile58.82%EvidenceCEvaluatedNov 17, 2026
BenchmarkLM Arena DocumentScore1,434.33 ratingRank30Participants39Percentile23.68%EvidenceAEvaluatedNov 9, 2026
BenchmarkLM Arena Vision Style ControlScore1,264.81 ratingRank30Participants113Percentile74.11%EvidenceAEvaluatedNov 9, 2026
BenchmarkAA SciCode SubtasksScore45.49%Rank45Participants89Percentile50.00%EvidenceBEvaluatedN/A
BenchmarkLM Arena TextScore1,442.54 ratingRank53Participants221Percentile76.36%EvidenceAEvaluatedNov 11, 2026
BenchmarkLM Arena Text FactualityScore1,449.05 ratingRank53Participants134Percentile60.90%EvidenceAEvaluatedNov 11, 2026
BenchmarkLM Arena Text Style ControlScore1,451.82 ratingRank57Participants221Percentile74.55%EvidenceAEvaluatedNov 11, 2026
BenchmarkAA GPQA DiamondScore85.66%Rank64Participants199Percentile68.18%EvidenceBEvaluatedN/A
BenchmarkHumanity's Last ExamScore26.50%Rank65Participants119Percentile45.76%EvidenceCEvaluatedNov 17, 2026
BenchmarkAA HLE Text No ToolsScore23.63%Rank75Participants200Percentile62.81%EvidenceBEvaluatedN/A
BenchmarkGPQAScore84.30%Rank78Participants265Percentile70.83%EvidenceCEvaluatedNov 17, 2026
BenchmarkLM Arena WebdevScore1,370.27 ratingRank82Participants111Percentile26.36%EvidenceAEvaluatedNov 15, 2026
BenchmarkAA CritPtScore1.43%Rank85Participants200Percentile57.79%EvidenceBEvaluatedN/A
BenchmarkAA LCR v1.1Score69.67%Rank91Participants193Percentile53.13%EvidenceBEvaluatedN/A
BenchmarkAA Omniscience AccuracyScore20.03%Rank126Participants201Percentile37.50%EvidenceBEvaluatedN/A

Gemma 4 31B Arena Results

Preference and agent-evaluation results for the default version.

30 of 88 rows
Columns

Show columns

Sort by
Arena
Category
Rank
Rating / score
Votes
Observations
Result date
ArenavisionCategorycaptioningRank19Rating / score1,209.14Votes103ObservationsN/AResult dateNov 9, 2026
ArenavisionCategorycreative writing visionRank19Rating / score1,286.42Votes2,847ObservationsN/AResult dateNov 9, 2026
ArenavisionCategoryhumorRank23Rating / score1,279.90Votes1,707ObservationsN/AResult dateNov 9, 2026
ArenavisionCategoryentity recognitionRank24Rating / score1,227.43Votes367ObservationsN/AResult dateNov 9, 2026
ArenavisionCategoryhomeworkRank24Rating / score1,309.68Votes6,717ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategorycreative writing visionRank24Rating / score1,265.88Votes2,847ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategorycaptioningRank25Rating / score1,161.48Votes103ObservationsN/AResult dateNov 9, 2026
Arenadocument style controlCategoryoverallRank26Rating / score1,452.04Votes15,431ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategoryentity recognitionRank26Rating / score1,197.30Votes367ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategoryhomeworkRank26Rating / score1,303.80Votes6,717ObservationsN/AResult dateNov 9, 2026
ArenavisionCategoryoverallRank28Rating / score1,279.39Votes41,787ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategoryhumorRank29Rating / score1,259.52Votes1,707ObservationsN/AResult dateNov 9, 2026
ArenadocumentCategoryoverallRank30Rating / score1,434.33Votes15,431ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategoryoverallRank30Rating / score1,264.81Votes41,787ObservationsN/AResult dateNov 9, 2026
ArenavisionCategoryocrRank31Rating / score1,285.02Votes29,112ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategorychineseRank32Rating / score1,305.93Votes2,769ObservationsN/AResult dateNov 9, 2026
ArenatextCategoryrussianRank33Rating / score1,460.31Votes824ObservationsN/AResult dateNov 11, 2026
ArenavisionCategorychineseRank33Rating / score1,326.58Votes2,769ObservationsN/AResult dateNov 9, 2026
ArenavisionCategorydiagramRank33Rating / score1,286.02Votes10,636ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategoryocrRank34Rating / score1,276.56Votes29,112ObservationsN/AResult dateNov 9, 2026
ArenavisionCategoryenglishRank36Rating / score1,272.46Votes16,288ObservationsN/AResult dateNov 9, 2026
Arenavision style controlCategorydiagramRank36Rating / score1,286.32Votes10,636ObservationsN/AResult dateNov 9, 2026
Arenatext style controlCategoryexpertRank37Rating / score1,498.04Votes501ObservationsN/AResult dateNov 11, 2026
Arenatext style controlCategoryrussianRank37Rating / score1,467.94Votes824ObservationsN/AResult dateNov 11, 2026
Arenavision style controlCategoryenglishRank37Rating / score1,257.44Votes16,288ObservationsN/AResult dateNov 9, 2026
ArenatextCategorymathRank42Rating / score1,462.41Votes430ObservationsN/AResult dateNov 11, 2026
Arenatext style controlCategoryinstruction followingRank46Rating / score1,453.14Votes1,801ObservationsN/AResult dateNov 11, 2026
Arenatext style controlCategorymathRank46Rating / score1,467.06Votes430ObservationsN/AResult dateNov 11, 2026
Arenatext style controlCategoryindustry mathematicalRank47Rating / score1,471.19Votes344ObservationsN/AResult dateNov 11, 2026
Arenatext style controlCategorylonger queryRank47Rating / score1,468.76Votes1,802ObservationsN/AResult dateNov 11, 2026

Gemma 4 31B Pricing

Official vendor API pricing appears first, followed by individual provider offers.

Official API
N/A
Official provider
N/A
Lowest third-party
From $0.09 input, $0.34 output per 1M via Adaption
Tracked offerings
39
30 of 39 rows
Columns

Show columns

Sort by
Provider
Provider model ID
Region
Input / 1M
Output / 1M
Context
Updated
ProviderNvidiaProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1MN/AOutput / 1MN/AContext256KUpdatedNov 18, 2026
ProviderGoogleProvider model IDgemma-4-31b-itRegionglobalInput / 1MN/AOutput / 1MN/AContext262.1KUpdatedNov 18, 2026
ProviderRequestyProvider model IDgemma-4-31b-itRegionglobalInput / 1MN/AOutput / 1MN/AContext262.1KUpdatedNov 18, 2026
ProviderPoeProvider model IDgoogle/gemma-4-31bRegionglobalInput / 1MN/AOutput / 1MN/AContext262.1KUpdatedNov 18, 2026
ProviderAMDProvider model IDGemma-4-31B-ITRegionglobalInput / 1MN/AOutput / 1MN/AContext262.1KUpdatedNov 18, 2026
ProviderAdaptionProvider model IDgemma-4-31b-itRegionglobalInput / 1M$0.09Output / 1M$0.34Context256KUpdatedNov 18, 2026
ProviderFastRouterProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.10Output / 1M$0.30Context262.1KUpdatedNov 18, 2026
ProviderDevPass (LLM Gateway)Provider model IDgemma-4-31b-itRegionglobalInput / 1M$0.10Output / 1M$0.25Context262.1KUpdatedNov 18, 2026
ProviderKilo GatewayProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.10Output / 1M$0.34Context262.1KUpdatedNov 18, 2026
ProviderCoreWeaveProvider model IDgoogle/gemma-4-31B-itRegionglobalInput / 1M$0.10Output / 1M$0.34Context262.1KUpdatedNov 18, 2026
ProviderKenariProvider model IDgemma-4-31b-itRegionglobalInput / 1M$0.1029Output / 1M$0.4057Context262.1KUpdatedNov 18, 2026
ProviderNanoGPTProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.11Output / 1M$0.33Context262.1KUpdatedNov 18, 2026
ProviderNanoGPTProvider model IDTEE/gemma4-31bRegionglobalInput / 1M$0.12Output / 1M$0.37Context262.1KUpdatedNov 18, 2026
ProviderDoublewordProvider model IDgoogle/gemma-4-31B-itRegionglobalInput / 1M$0.12Output / 1M$0.35Context262.1KUpdatedNov 18, 2026
ProviderOrcaRouterProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.13Output / 1M$0.38Context262.1KUpdatedNov 18, 2026
ProviderHugging FaceProvider model IDgoogle/gemma-4-31B-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderCrusoeProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderMerge GatewayProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderOllama CloudProvider model IDgemma4:31bRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderFriendliProvider model IDgoogle/gemma-4-31B-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderVercel AI GatewayProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderNovita AIProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderAbacusProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.14Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderNeuralwattProvider model IDgemma-4-31bRegionglobalInput / 1M$0.144Output / 1M$0.42Context262.1KUpdatedNov 18, 2026
ProviderKelnProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.15Output / 1M$0.46Context262.1KUpdatedNov 18, 2026
ProviderPhalaProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.15Output / 1M$0.46Context262.1KUpdatedNov 18, 2026
ProviderDigitalOceanProvider model IDgemma-4-31B-itRegionglobalInput / 1M$0.18Output / 1M$0.50Context256KUpdatedNov 18, 2026
ProviderDeep InfraProvider model IDgoogle/gemma-4-31B-itRegionglobalInput / 1M$0.20Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
ProviderOpenRouterProvider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.20Output / 1M$0.40Context262.1KUpdatedNov 18, 2026
Providerai&Provider model IDgoogle/gemma-4-31b-itRegionglobalInput / 1M$0.20Output / 1M$0.50Context262.1KUpdatedNov 18, 2026

Gemma 4 31B Runtime Performance

Provider-specific output speed and catalog latency for Gemma 4 31B. Runtime does not affect the capability score.

No runtime data

No provider-specific speed or latency record is linked to the default version yet.

Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.

Gemma 4 31B Specifications

Technical details for the model's default version.

Version
Gemma 4 31B
Released
Apr 2, 2026
Knowledge cutoff
Jan 1, 2025
Parameters
30.7B
Context window
262.1K
Max output
262.1K
Inputs
audio, image, text, video
Outputs
text
Open weights
Yes
License
Apache 2.0

Gemma 4 31B Versions

Available versions of this model. The score column identifies the version used in the overall ranking.

1 row
Columns

Show columns

Sort by
Version
Released
LLMBoard
Parameters
Context
Max output
Open weights
License
VersionGemma 4 31BReleasedApr 2, 2026LLMBoard57.34Parameters30.7BContext262.1KMax output262.1KOpen weightsYesLicenseApache 2.0

Models similar to Gemma 4 31B

Recommendations prioritize the same model type and family, then the closest LLMBoard score.

#122-0.59
GO

Gemini 3.5 Flash Cyber

Google

56.75 LLMBoard

Details
#106+2.47
GO

Gemini 3.8 Flash Cyber

Google

59.81 LLMBoard

Details
#138-5.66
GO

Gemini 3.5 Flash Lite

Google

51.68 LLMBoard

Details
#96+5.84
GO

Gemini 3 Flash

Google

63.18 LLMBoard

Details
#148-7.23
GO

Gemma 4 26B A4B

Google

50.11 LLMBoard

Details
#150-7.63
GO

Gemini 3.1 Flash Lite

Google

49.71 LLMBoard

Details

What is Gemma 4 31B?

Key information about Gemma 4 31B and its available data.

Gemma 4 31B is a large language model from Google based on the same research as Gemini 3. It uses Per-Layer Embeddings, Shared KV Cache, alternating sliding-window and global attention, and variable aspect ratio vision encoding.

Data as of 2026-11-17.

FAQ

Common questions about Gemma 4 31B.

When was Gemma 4 31B released?

Gemma 4 31B's default version was released on Apr 2, 2026.

How much does Gemma 4 31B cost?

No official standard PAYG price is currently available for Gemma 4 31B. The lowest tracked third-party offer starts at $0.09 input and $0.34 output via Adaption.

Who created Gemma 4 31B?

Gemma 4 31B was created by Google.

What is the context window for Gemma 4 31B?

The default version has a 262.1K token context window.

Is Gemma 4 31B open weight?

Yes. The default version is marked as open weight under Apache 2.0.

How many API providers offer Gemma 4 31B?

39 provider offerings are linked to the default version.

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

Browse runtime rankings