Google
Google: Gemma 4 26B A4B (free)
Input / M
Routed
Output / M
Context
Max Output
33K
Usage
Unranked
Quality
Unscored
Providers
2
Best Uptime 30m
100.00%
Latency (p50)
1039 ms
Throughput
39.0 tok/s
Released
April 3, 2026
Modalities
image, text, video
About
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.
Capabilities
Vision
Tools
Reasoning
Streaming
JSON Mode
Host offerings
Per-host pricing, uptime, latency
No host data yet — will populate on next sync.
Price History
0 recorded changes
No price changes recorded yet.
Activity Timeline
- No events yet.
Versus
Nearest peers by price, context, and capabilities. One click to compare.
No comparable models found.
Show this WM Score
Free live badge for your README, docs or site
Live badge — it always shows the current WM Score. Free to use anywhere, no key needed.
Markdown
[](https://whatmodel.app/models/b8d6beed-e7e9-4e55-bc0f-0fb4f69ac213)HTML
<a href="https://whatmodel.app/models/b8d6beed-e7e9-4e55-bc0f-0fb4f69ac213"><img src="https://whatmodel.app/api/public/embed/score/b8d6beed-e7e9-4e55-bc0f-0fb4f69ac213" alt="Google: Gemma 4 26B A4B (free) WM Score" height="26" /></a>Image URL
https://whatmodel.app/api/public/embed/score/b8d6beed-e7e9-4e55-bc0f-0fb4f69ac213First seen 43d ago · Last verified 5m ago