πŸ”“ **Step 3.7 Flash free for 30 days**

Nous Research has opened 30-day free access to StepFun's multimodal MoE model, Step 3.7 Flash, through the Nous Portal. The model features 198 billion parameters (11 billion active) and supports a context window of up to 256K tokens. It is optimized for agentic tasks: coding, searching, and working with media content.

🌍 The availability of powerful multimodal MoE models accelerates the creation of autonomous agents. The architecture provides a balance between efficiency and performance.

πŸ‘€ You can now test this cutting-edge model for agentic tasks in real-world scenarios for free via Nous Research.

Source 1: https://developer.nvidia.com/blog/run-step-3-7-flash-on-nvidia-gpus-with-enterprise-ready-multimodal-ai/ Source 2: https://platform.stepfun.ai/docs/en/guides/models/reasoning