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...
Tested on 4 benchmarks with 0.0% average. Top scores: Chatbot Arena Elo — Overall (1437.9%), Artificial Analysis — Agentic Index (32.1%), Artificial Analysis — Quality Index (31.2%).
Chatbot Arena overall Elo rating. Crowdsourced human preference ranking from blind head-to-head comparisons across all topics.
Artificial Analysis Agentic Index. Composite score measuring agent capability across tool use and planning tasks.
Artificial Analysis Quality Index. Composite quality score combining multiple benchmark results into a single metric.
Artificial Analysis Coding Index. Composite coding quality score from multiple code benchmarks.
vs Qwen VL Max
vs Elephant
vs Trinity Large Thinking- Typemultimodal
- Context262K tokens (~131 books)
- ReleasedApr 2026
- LicenseOpen Source
- StatusActive
- Cost / Message~$0.000
Google DeepMindFree