Google

Gemma 4 31B IT

CTX262K

63Quality

Reasoning · Vision · Tools

Downloads9M/mo

Likes4.2K

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

Open weightsApache 2.0852/1044

Drag to turn

GoogleTextPromo

Gemma 4 31B IT

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

gemma-4-31b-it

Type: Text
Quality: 63 · Top 37% on the Epoch Capabilities Index
Context window: 262,144 tokens
Max output: 32,768 tokens
Input price: —
Output price: —
Cache read: —
Cache write: —
Parameters: 31B (31,273,088,876)
Downloads: 9M in the last 30 days
Likes: 4.2K
Access to weights: Open download
Released: Apr 2, 2026
Knowledge cutoff: —
Takes in: Text, Image
Puts out: Text
Reasoning: Yes
Tool calling: Yes
Structured output: Yes
File attachments: Yes
Access: Open weights
License: Apache 2.0
Rarity: Promo
Set number: 852/1044

Open Google docs Weights on Hugging Face Quality 63 Where to run it 44 Evolution 4/4 Race it 30 tok/s Meet its creature Battle Qwen3 8B

License read from the model’s Hugging Face card. Specs from models.dev, downloads from Hugging Face, last changed Oct 19, 2026. Official source: ai.google.dev.

How good is it?

Public leaderboard scores for Gemma 4 31B IT, refreshed daily from Epoch AI and LMArena.

Top 37% of the 277 models on the Epoch Capabilities Index: #102.

Quality is the share of a leaderboard’s models this one beats, from 0 to 100. Chat models are rated on Epoch AI’s Capabilities Index when it lists them, otherwise on LMArena; image and video models on LMArena’s image and video arenas.

Leaderboards

Epoch Capabilities Index#102 of 277
One number fitted from dozens of hard benchmarks, from science questions to agentic coding.142.9 index
LMArena Text#67 of 416
People compare two anonymous answers and vote for the better one.1,443 rating
Listed as gemma-4-31b
LMArena Text · Coding#85 of 411
The text arena’s coding prompts only.1,459 rating
Listed as gemma-4-31b
LMArena Text · Math#54 of 400
The text arena’s math prompts only.1,464 rating
Listed as gemma-4-31b
LMArena Text · Creative writing#75 of 414
The text arena’s creative writing prompts only.1,415 rating
Listed as gemma-4-31b
LMArena WebDev#107 of 145
Each model builds a web app from the same prompt; people pick the better one.1,368 rating
Listed as Gemma 4 31B
LMArena Vision#38 of 165
Questions about images, voted by people.1,277 rating
Listed as gemma-4-31b

Benchmarks Epoch AI runs

Scores from Epoch AI and LMArena, licensed CC BY 4.0. Boards list models under their own names, so a close match is shown with the name it was scored under. Credits

Where to run it

41 providers sell Gemma 4 31B IT through an API, and 3 Hugging Face hosts serve it. Prices are per million tokens, cheapest first.

CheapestDevPass (LLM Gateway)$0.10 in · $0.25 out
Free to try8 providersBothub, Kenari, Nvidia, …
From the labGoogle— in · — out
Fastest on Hugging FaceNovita30 tokens/s

API providers

Provider	Input	Output	Context
BothubFree`gemma-4-31b-it:free`	Free	Free	262K
KenariFree`gemma-4-31b-it`	Free	Free	262K
NvidiaFree`google/gemma-4-31b-it`	Free	Free	256K
OpenRouterFree`google/gemma-4-31b-it:free`	Free	Free	262K
PoeFree`google/gemma-4-31b`	Free	Free	262K
QVACFree`gemma4-31b`	Free	Free	262K
RequestyFree`gemma-4-31b-it`	Free	Free	262K
UnoRouterFree`gemma-4-31b-it:free`	Free	Free	262K
DevPass (LLM Gateway)Cheapest paid`gemma-4-31b-it`	$0.10	$0.25	262K
LLM Gateway`consensusprotocol/gemma-4-31b-it`	$0.10	$0.25	262K

Show 31 more providers

Provider	Input	Output	Context
CrofAI`gemma-4-31b-it`	$0.10	$0.30	262K
CoreWeave`google/gemma-4-31B-it`	$0.10	$0.34	262K
Kilo Gateway`google/gemma-4-31b-it`	$0.10	$0.34	262K
NanoGPT`google/gemma-4-31b-it:thinking`	$0.10	$0.34	262K
Lilac`google/gemma-4-31b-it`	$0.11	$0.35	262K
FastRouter`google/gemma-4-31b-it`	$0.13	$0.38	262K
OrcaRouter`google/gemma-4-31b-it`	$0.13	$0.38	262K
Cortecs`gemma-4-31b-it`	$0.22	$0.39	262K
SiliconFlow`google/gemma-4-31B-it`	$0.13	$0.40	262K
Abacus`google/gemma-4-31b-it`	$0.14	$0.40	262K
Amazon Bedrock`google.gemma-4-31b`	$0.14	$0.40	262K
Crusoe`google/gemma-4-31b-it`	$0.14	$0.40	262K
Friendli`google/gemma-4-31B-it`	$0.14	$0.40	262K
Merge Gateway`google/gemma-4-31b-it`	$0.14	$0.40	262K
Novita AI`google/gemma-4-31b-it`	$0.14	$0.40	262K
Ollama Cloud`gemma4:31b`	$0.14	$0.40	262K
Vercel AI Gateway`google/gemma-4-31b-it`	$0.14	$0.40	262K
Deep Infra`google/gemma-4-31B-it`	$0.20	$0.40	262K
Neuralwatt`gemma-4-31b`	$0.14	$0.42	262K
DigitalOcean`gemma-4-31B-it`	$0.18	$0.50	256K
ai&`google/gemma-4-31b-it`	$0.20	$0.50	262K
Infomaniak`google/gemma-4-31B-it`	$0.25	$0.50	100K
Pioneer`google/gemma-4-31B-it`	$0.50	$0.50	33K
Berget.AI`google/gemma-4-31B-it`	$0.28	$0.55	128K
Tinfoil`gemma4-31b`	$0.40	$1.00	262K
SambaNova`gemma-4-31B-it`	$0.38	$1.15	131K
AI/ML API`google/gemma-4-31b-it`	$1.36	$2.05	256K
Regolo AI`gemma4-31b`	$0.46	$2.42	100K
Opper`gemma-4-31b-it`	$0.46	$2.44	256K
GoogleOfficial`gemma-4-31b-it`	—	—	262K
Tempr Gateway`google/gemma-4-31b-it`	—	—	262K

Hugging Face Inference Providers

Host	Input	Output	Context	Speed	First token
NovitaFastestTools	$0.14	$0.40	262K	30 tok/s	0.6s
DeepInfraTools	$0.20	$0.40	262K	14 tok/s	1.7s
Featherless	—	—	—	—	—

From models.dev and Hugging Face, checked daily. Resellers’ listings can lag behind their price pages, so confirm with the provider before you commit.

Gemma 4 31B IT

Gemma 4 31B IT

How good is it?

Leaderboards

Benchmarks Epoch AI runs

Where to run it

API providers

Hugging Face Inference Providers

More from Google

Gemini 4 Argon

embeddinggemma-2

Gemini 3.8 Flash

timesfm-3.0-pytorch

Gemini 3.7 Flash

Gemini 3.6 Flash

Gemini 3.5 Flash Lite

Nano Banana 2 Lite