← All models

Google: Gemma 4 26B A4B

google/gemma-4-26b-a4b-it

VisionTool useJSONReasoningStreaming

Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...

Pricing

Input

$0.06 / 1M

Output

$0.33 / 1M

Specs

Context

262,144 tokens

Input

image, text, video

Output

text

Released: 2026-04

Supported parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p

Open weights · HuggingFace

11,164,296 downloads/mo
1,043 likes
apache-2.0 image-text-to-text
View on HuggingFace →