← Back to all models
💬
Google: Gemma 4 26B A4B (free)
🔥 50trending
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...
#text+image+video->text#top-provider
🧮
Undisclosed
Parameters
📏
262K tokens
Context Window
🔒
Proprietary
License
📅
Apr 3, 2026
Released
🔌
API Available
This model is accessible via API for integration into your applications.
⭐ Related Models
💬
Claude 3.5 Sonnet
Anthropic
🔥 85
The model that defined a generation. Fast, smart, and incredibly capable across coding, analysis, and creative tasks.
🔒 proprietary📏 200K ctx💰 $3
Oct 22, 2024View details →
💬
Claude 3.5 Haiku
Anthropic
🔥 72
Ultra-fast and cost-effective. Best for high-volume tasks where speed matters more than peak intelligence.
🔒 proprietary📏 200K ctx💰 $0.80
Nov 4, 2024View details →
💬
GPT-4o Mini
OpenAI
🔥 70
Compact and affordable. Surprisingly capable for its price point, ideal for high-volume applications.
🔒 proprietary📏 128K ctx💰 $0.15
Jul 18, 2024View details →