Google DeepMind's open model family for developers, including multimodal releases.
Pricing
Model profile · Current snapshot
Input / 1M
$0.11
Output / 1M
$0.25
Blended / 1M
$0.145
Output TPS
0
TTFT (s)
0
Model Intelligence
Arena ranking
5
Benchmarkable
No
Model level
family
Intelligence Index
4.8
Math Index
20.7
MMLU Pro
0.67
GPQA
0.43
HLE
0.05
LiveCodeBench
0.14
SciCode
0.21
MATH-500
0.88
AIME
0.25
AIME 2025
0.21
IFBench
0.32
LCR
0.06
TerminalBench Hard
0.04
TAU2
0.11
Recent stories
2 linked stories
releasePRIMARY2026-06-03
Gemma 4 12B releases with 256K context and unified audio-vision input
Google’s new Gemma 4 12B ships as an encoder-free open model for text, image, audio, and video tasks with a 256K context window. Early GGUF ports and local benchmarks make it a plausible on-device multimodal option for creator tooling and experimentation.
releasePRIMARY2026-04-02
Google DeepMind releases Gemma 4 under Apache 2.0 with 31B Dense, 26B MoE, and 256K context
Google DeepMind shipped four Gemma 4 models with multimodal input, including 31B Dense, 26B MoE, and two edge variants available through AI Studio, Hugging Face, Kaggle, and Ollama. Early community tests say local performance and usable context windows still vary by runtime, quantization, and GPU memory.