Skip to main content
Back to Explore

Google: Gemma 4 31B

Google·Released 2026-04-02
Open Source31B262K ctxApache 2.0Multimodal

About

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...

Pricing

Input

$0.12

per 1M tokens

Output

$0.35

per 1M tokens

Blended

$0.18

per 1M tokens

Cheaper than 74% of models. Median price is $0.67/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.18

Monthly

$5.32

vs. Similar Models

DeepSeek V4 Flash (Reasoning, High Effort)
$0.17-1%
MiMo-V2.5
$0.17-1%
DeepSeek V4 Flash (Non-reasoning)
$0.17-1%
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)
$0.17-1%

Performance

Context Window

262K

tokens

Larger than 60% of models

Max Output

262K

tokens

100% of context

Context Window Comparison

MoonshotAI: Kimi K2.6
262KSame
MoonshotAI: Kimi K2.7 Code
262KSame
Nex AGI: Nex-N2-Pro
262KSame
apache-2.031BGGUF / GPTQ / AWQ
Downloads

803.9K

Likes

458

VRAM (FP16)

24-48 GB

GPU

A6000 / M3 Ultra

Quick Compare

Similar Models

Compare all 7 models