llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsGemma 3 12B

Google model product

Gemma 3 12B

Gemma 3 12B is a 12-billion-parameter vision-language model from Google, handling text and image input and generating text output. It features a 128K context window, multilingual support, and open weights. Suitable for question answering, summarization, reasoning, and image understanding tasks.

Updated Aug 10, 2026. Default version: Gemma 3 12B

Compare
LLMBoard score39.2Gemma 3 12B
Coverage80%15 benchmark families
Context window131.1KTokens
Official input priceN/AOfficial price unavailable

On this page

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

Model specification

Structured fields from the published default version.

Version
Gemma 3 12B
Released
Mar 12, 2025
Knowledge cutoff
Unknown
Parameters
12B
Context window
131.1K
Max output
131.1K
Inputs
image, text
Outputs
text
Open weights
No
License
Gemma

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

Gemma 3 12B category scores

Benchmark results

Published benchmark records for the scored version Gemma 3 12B.

26 rows
Columns

Show columns

VQAv2 (val)0.713100.0%CAug 7, 2026
ECLeKTic0.13871.4%CAug 7, 2026
Bird-SQL (dev)0.54750.0%CAug 7, 2026
HiddenMath0.541375.0%CAug 7, 2026
Natural2Code0.84857.1%CAug 7, 2026
BIG-Bench Hard0.962175.0%CAug 7, 2026
FACTS Grounding0.861358.3%CAug 7, 2026
Global-MMLU-Lite0.771453.9%CAug 7, 2026
BIG-Bench Extra Hard0.281130.0%CAug 7, 2026
InfoVQA0.68912.5%CAug 7, 2026
MMMU (val)0.6101110.0%CAug 7, 2026
TextVQA0.7131514.3%CAug 7, 2026
MATH0.8147181.4%CAug 7, 2026
WMT24++0.5152336.4%CAug 7, 2026
GSM8k0.9164868.1%CAug 7, 2026
DocVQA0.9222616.0%CAug 7, 2026
IFEval0.9226567.2%CAug 7, 2026
MathVista-Mini0.622234.5%CAug 7, 2026
MBPP0.7223741.7%CAug 7, 2026
ChartQA0.823244.3%CAug 7, 2026
AI2D0.8243225.8%CAug 7, 2026
HumanEval0.9387650.7%CAug 7, 2026
SimpleQA0.142468.9%CAug 7, 2026
LiveCodeBench0.2657311.1%CAug 7, 2026
MMLU-Pro0.610312920.3%CAug 7, 2026
GPQA0.420223313.4%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

46 rows
Columns

Show columns

textgerman1331370.7166N/AAug 6, 2026
text style controlgerman1421371.2166N/AAug 6, 2026
textindustry business and management and financial operations1451376.3364N/AAug 6, 2026
textindustry legal and government1491381.2217N/AAug 6, 2026
text style controlindustry business and management and financial operations1531385.7364N/AAug 6, 2026
textcreative writing1641331.8600N/AAug 6, 2026
text style controlindustry legal and government1671382.1217N/AAug 6, 2026
text style controlrussian1691358.0285N/AAug 6, 2026
textindustry life and physical and social science1711365.9760N/AAug 6, 2026
textrussian1751335.8285N/AAug 6, 2026
text style controlcreative writing1761333.4600N/AAug 6, 2026
text style controlnon english1851336.61,499N/AAug 6, 2026
textnon english1861318.61,499N/AAug 6, 2026
text style controlindustry life and physical and social science1891367.8760N/AAug 6, 2026
text style controlmulti turn1891344.1471N/AAug 6, 2026
textindustry mathematical1901335.5333N/AAug 6, 2026
textexclude ties1921289.42,526N/AAug 6, 2026
textindustry medicine and healthcare1921334.1155N/AAug 6, 2026
text style controlindustry mathematical1921349.2333N/AAug 6, 2026
textoverall1931334.33,829N/AAug 6, 2026
textindustry entertainment and sports and media1931298.2697N/AAug 6, 2026
textindustry writing and literature and language1941307.31,010N/AAug 6, 2026
textmulti turn1941333.6471N/AAug 6, 2026
textenglish1971348.12,330N/AAug 6, 2026
text style controlindustry entertainment and sports and media1971306.2697N/AAug 6, 2026
text style controllonger query1971345.8371N/AAug 6, 2026
textlonger query2011316.9371N/AAug 6, 2026
text style controlmath2041317.6389N/AAug 6, 2026
textinstruction following2051299.71,145N/AAug 6, 2026
textmath2051307.2389N/AAug 6, 2026
text style controloverall2051342.03,829N/AAug 6, 2026
text style controlexclude ties2051296.12,526N/AAug 6, 2026
text style controlinstruction following2081321.31,145N/AAug 6, 2026
text style controlindustry writing and literature and language2091313.51,010N/AAug 6, 2026
texthard prompts2101309.6977N/AAug 6, 2026
text style controlenglish2141348.82,330N/AAug 6, 2026
texthard prompts english2151318.4700N/AAug 6, 2026
textindustry software and it services2191313.7867N/AAug 6, 2026
textexpert2221247.3186N/AAug 6, 2026
text style controlhard prompts2241332.0977N/AAug 6, 2026
text style controlhard prompts english2351337.8700N/AAug 6, 2026
text style controlindustry medicine and healthcare2371322.9155N/AAug 6, 2026
textcoding2431281.7543N/AAug 6, 2026
text style controlindustry software and it services2491334.7867N/AAug 6, 2026
text style controlexpert2541271.7186N/AAug 6, 2026
text style controlcoding2631316.3543N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
N/A
Official provider
N/A
Lowest third-party
From $0.05 input, $0.10 output per 1M via Helicone
Tracked offerings
4
4 rows
Columns

Show columns

Heliconegemma-3-12b-itglobal$0.05$0.10131.1KAug 7, 2026
NovitaAIgoogle/gemma-3-12b-itglobal$0.05$0.10131.1KAug 7, 2026
OpenRoutergoogle/gemma-3-12b-itglobal$0.05$0.15131.1KAug 7, 2026
Neongemma-3-12bglobal$0.15$0.50131.1KAug 7, 2026

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

Gemma 3 12B versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

Gemma 3 12BMar 12, 202539.212B131.1K131.1KNoGemma

What is Gemma 3 12B?

A concise description based on the published model registry.

Gemma 3 12B is a 12-billion-parameter vision-language model from Google, handling text and image input and generating text output. It features a 128K context window, multilingual support, and open weights. Suitable for question answering, summarization, reasoning, and image understanding tasks.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

Gemma 3 12B vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

Gemma 3 12BvsHermes 3 70BGemma 3 12BvsClaude Sonnet 3Gemma 3 12BvsQwen3 30B A3BGemma 3 12BvsJamba 1.5 LargeGemma 3 12BvsDiffusionGemma 26B A4BGemma 3 12BvsDeepSeek-V3

Models similar to Gemma 3 12B

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

#118-0.9
GO

DiffusionGemma 26B A4B

Google

38.3 LLMBoard

DetailsCompare
#122-2.3
GO

Gemini 1.5 Flash

Google

36.9 LLMBoard

DetailsCompare
#96+6.0
GO

Gemma 3 27B

Google

45.1 LLMBoard

DetailsCompare
#93+7.9
GO

Gemma 2 27B

Google

47.1 LLMBoard

DetailsCompare
#135-8.4
GO

Gemini 2.5 Flash Lite

Google

30.7 LLMBoard

DetailsCompare
#91+9.0
GO

Gemma 4 12B

Google

48.1 LLMBoard

DetailsCompare

FAQ

Common questions about Gemma 3 12B.

When was Gemma 3 12B released?

Gemma 3 12B's default version was released on Mar 12, 2025.

How much does Gemma 3 12B cost?

No official standard PAYG price is currently available for Gemma 3 12B. The lowest tracked third-party offer starts at $0.05 input and $0.10 output via Helicone.

Who created Gemma 3 12B?

Gemma 3 12B is published under Google in the model registry.

What is the context window for Gemma 3 12B?

The default version has a 131.1K token context window.

Is Gemma 3 12B open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer Gemma 3 12B?

4 published provider offerings are linked to the default version.

What models should I compare Gemma 3 12B with?

Nearby ranked alternatives include Hermes 3 70B, Claude Sonnet 3, Qwen3 30B A3B.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai