Skip to content

DiffusionGemma 26B A4B

Disabled since Aug 19, 2026

Chat completions return empty content for this discrete-diffusion model under normal chat token budgets. It is unlisted rather than advertised as a live chat model.

google/diffusiongemma-26b-a4b-it

Vision · 262K context · Jun 2026 · 2 upstreams

DiffusionGemma 26B A4B is Google DeepMind's open-weights multimodal model that generates text via discrete diffusion. It builds on Gemma 4 26B A4B MoE with 25.2B total and 3.8B active parameters, a 256K context window, a thinking mode and function calling.

context
262K
input / 1M
$0
output / 1M
$0
modalities
text · image · video
speed, 24h
No data yet

Upstreams

2
  1. 1NVIDIAPlatformFreeFree
  2. 2NVIDIABYOKBilled to your key

Compare

Compared with models of similar price and context window.

ModelIn / out per 1MContext
DiffusionGemma 26B A4B$0 / $0262K
Agents A1$0 / $0262K
Apodex 1.1 Mini$0 / $0262K
Intern S2$0 / $0262K
Flint$0 / $0262K

Details

Owner
Google
Kind
Vision
Released
Jun 2026
Context
262K
Capabilities
Data policy
Provider retention
Endpoint
/api/v1/chat/completions