Google logo

Gemma 4 31B IT

google/gemma-4-31b-it

Largest Gemma 4 instruction model for open, self-hosted chat and reasoning

Source-linkedgemmaOpen weightsReleased 2026-04-02
Report data
Context window262.144K
Maximum output32.768K
Provider offers24
Knowledge cutoffUndisclosed
Capabilities

What this model supports

ReasoningYes
Tool callingYes
Structured outputYes
AttachmentsYes
Temperature controlYes
Open weightsYes
Creator
Google
Family
gemma
Modalities
Image, Text
License
Not documented
Release date
2026-04-02
Model ID
google/gemma-4-31b-it
Market availability

Provider pricing

Report price →
Providers offering Gemma 4 31B IT
ProviderProvider model IDContextInput / 1MOutput / 1MCached / 1MFeaturesAll pricing
Googlegemma-4-31b-it262.144KR T S
View
{}
Merge Gatewaygoogle/gemma-4-31b-it262.144KR T S
View
{}
Kenarigemma-4-31b-it262.144K$0.000$0.000R T S
View
{
  "input": 0,
  "output": 0
}
UnoRoutergemma-4-31b-it:free262.144K$0.000$0.000R T S
View
{
  "input": 0,
  "output": 0
}
Nvidiagoogle/gemma-4-31b-it256K$0.000$0.000R T ·
View
{
  "input": 0,
  "output": 0
}
NanoGPTgoogle/gemma-4-31b-it262.144K$0.100$0.350R · S
View
{
  "input": 0.1,
  "output": 0.35
}
CrofAIgemma-4-31b-it262.144K$0.100$0.300$0.020R T S
View
{
  "cache_read": 0.02,
  "input": 0.1,
  "output": 0.3
}
Lilacgoogle/gemma-4-31b-it262.1K$0.110$0.350R T S
View
{
  "input": 0.11,
  "output": 0.35
}
OpenRoutergoogle/gemma-4-31b-it262.144K$0.120$0.350$0.090R T S
View
{
  "cache_read": 0.09,
  "input": 0.12,
  "output": 0.35
}
LLM Gatewaygemma-4-31b-it262.144K$0.130$0.380R T S
View
{
  "input": 0.13,
  "output": 0.38
}
NEAR AI Cloudgoogle/gemma-4-31B-it262.144K$0.130$0.400$0.026R T S
View
{
  "cache_read": 0.026,
  "input": 0.13,
  "output": 0.4
}
OrcaRoutergoogle/gemma-4-31b-it262.144K$0.130$0.380R T S
View
{
  "input": 0.13,
  "output": 0.38
}
SiliconFlowgoogle/gemma-4-31B-it262.144K$0.130$0.400· T S
View
{
  "input": 0.13,
  "output": 0.4
}
Deep Infragoogle/gemma-4-31B-it262.144K$0.130$0.380R T S
View
{
  "input": 0.13,
  "output": 0.38
}
FastRoutergoogle/gemma-4-31b-it262.144K$0.130$0.380R T S
View
{
  "input": 0.13,
  "output": 0.38
}
Vercel AI Gatewaygoogle/gemma-4-31b-it262.144K$0.140$0.400· T S
View
{
  "input": 0.14,
  "output": 0.4
}
Hugging Facegoogle/gemma-4-31B-it262.144K$0.140$0.400R T S
View
{
  "input": 0.14,
  "output": 0.4
}
Abacusgoogle/gemma-4-31b-it262.144K$0.140$0.400R T S
View
{
  "input": 0.14,
  "output": 0.4
}
Friendligoogle/gemma-4-31B-it262.144K$0.140$0.400R T S
View
{
  "input": 0.14,
  "output": 0.4
}
Kilo Gatewaygoogle/gemma-4-31b-it262.144K$0.140$0.400R T ·
View
{
  "input": 0.14,
  "output": 0.4
}
NovitaAIgoogle/gemma-4-31b-it262.144K$0.140$0.400R T S
View
{
  "input": 0.14,
  "output": 0.4
}
Tinfoilgemma4-31b256K$0.400$1.000R T S
View
{
  "input": 0.4,
  "output": 1
}
Pioneergoogle/gemma-4-31B-it32.768K$0.500$0.500$0.500R T S
View
{
  "cache_read": 0.5,
  "cache_write": 0.5,
  "input": 0.5,
  "output": 0.5
}
Cerebrasgemma-4-31b131.072K$0.990$1.490R T S
View
{
  "input": 0.99,
  "output": 1.49
}
Evidence

Published benchmark results

Agentic Index vs top 8 models
GPT-5.6 Sol
54.0
Claude Fable 5
52.8
GPT-5.6 Terra
47.4
Anthropic: Claude Opus 4.8
47.2
Claude Sonnet 5
46.7
xAI: Grok 4.5
45.7
GPT-5.6 Luna
45.6
Gemma 4 31B IT
14.4
Coding Index vs top 8 models
GPT-5.6 Sol
77.4
GPT-5.6 Terra
76.7
Claude Fable 5
76.5
GPT-5.5
74.9
Anthropic: Claude Opus 4.8
74.3
Anthropic: Claude Opus 4.7
73.6
xAI: Grok 4.5
72.4
Gemma 4 31B IT
43.4
Intelligence Index vs top 8 models
Claude Fable 5
59.9
GPT-5.6 Sol
58.9
Anthropic: Claude Opus 4.8
55.7
GPT-5.6 Terra
55.0
GPT-5.5
54.8
xAI: Grok 4.5
53.8
Anthropic: Claude Opus 4.7
53.5
Gemma 4 31B IT
29.4