google/gemma-4-31b-it

Not Available

Gemma 4 31B

Name: Gemma 4 31B
Brand: Google

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...

4/2/2026

262,144 tokens

#22 Text (spanish)

Specifications

Modalities

Input

image

text

video

Output

text

Supported Parameters

frequency_penalty

include_reasoning

logit_bias

logprobs

max_tokens

min_p

presence_penalty

reasoning

repetition_penalty

response_format

seed

stop

structured_outputs

temperature

tool_choice

tools

top_a

top_k

top_logprobs

top_p

Max Output Tokens

262,144

Leaderboard

Text

🏆OverallELO: 1,451

#57

🇨🇳ChineseELO: 1,477

#79

🇬🇧EnglishELO: 1,460

#60

frenchELO: 1,462

#68

spanishELO: 1,466

#22

russianELO: 1,468

#34

💻CodingELO: 1,498

#60

🧮MathELO: 1,471

#32

✍️Creative WritingELO: 1,421

#65

📝Instruction FollowingELO: 1,453

#43

🌶️Hard PromptsELO: 1,473

#60

💬Multi-TurnELO: 1,464

#57

Vision

🏆OverallELO: 1,257

#33

Code

🏆OverallELO: 1,363

#72