Gemma 4 26B A4B It (free)
Google · text, image, video → text
Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model developed by Google DeepMind. Featuring an expansive context length of 262,144 tokens, this model is built to handle highly complex and long-form tasks. While containing 25.2B total parameters, only 3.8B activate per token during inference, delivering premium quality with outstanding efficiency.
Input$0 /M
Output$0 /M