Google

Gemma 4 31B

by Googletext+image+video->text14 endpoints
Neura Intelligence Index
66.4/ 100
Rank
#109
Confidence
medium
Benchmarks
5(13% coverage)

Overview

Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...

Capabilities

Text generation
Image understanding
Video understanding
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities

Modalities

Input
ImageTextVideo
Output
Text

Technical Specifications

Context Window
262.1K tokens
Max Output
16.4K tokens
Tokenizer
Gemma
Uptime (24h)
9986.4%
Parameters
30700000000B
License
apache_2_0Open Weights
Country
🇺🇸US
Release Date
April 2, 20265mo

Supported Parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Benchmark Performance

general

BenchmarkScoreSource
MMMLU
88.4
verified
MRCR v2
66.4
verified

reasoning

BenchmarkScoreSource
GPQA Diamond
84.3
verified
Humanity's Last Exam
26.5
verified

multimodal

BenchmarkScoreSource
MMMU-Pro
76.9
verified

Performance Over Time

Pricing

Per 1M tokens
Input$0.09
Output$0.34
Cache read$0.05
Added
April 2, 2026
Last synced 9/18/2026