#22Google DeepMindOpen weights

Gemma 4 26B A4B

Google's 26B mixture-of-experts model with 3.8B active parameters per token. Scores 17 on the Artificial Analysis Intelligence Index at $0.10 per 1M input and $0.37 per 1M output tokens. Apache 2.0.

Overall
41.1
Input $/M
$0.10
Output $/M
$0.37
Context
262K
Released
Apr 2, 2026
Updated
Oct 3, 2026

Category scores

Overall rank history

Weekly snapshots. Lower is better.

Strengths

  • Cheapest small model at $0.10 per 1M input tokens
  • Open weights (Apache 2.0)
  • Best cost-efficiency in the small-model class (88.3)

Weaknesses

  • Lowest intelligence score among the ranked small models
  • Smaller 262k context window