Nextbit

Open-source 1.6T-parameter MoE with 49B active, built for advanced reasoning and long-horizon agents.

Efficient Mixture-of-Experts Gemma 4 with only 4B active parameters per token.

Efficient vision-language MoE with linear attention and 3B active parameters.

OpenAI's compact open-weights model for efficient inference.

Efficient 30B MoE language model with 3B active parameters per token.

Mid-size dense Qwen3 model balancing capability and efficiency.

Microsoft's compact research model with strong STEM reasoning.

Second-generation open model from Google DeepMind.