Our first model: Naive-N0.5-Flash

AI at the core of building the next generation of models.

Up to 2,000 tokens per second for high-speed inference.

Frontier intelligence, open to build on.

Naive-N0.5-Flash: Building Frontier AI with AI

An open-weight 309B MoE model for coding and AI R&D, with native 1M context, no full attention, and AI-optimized inference up to 2,000 tokens/s.

Read the article

Get started

API Platform

Access the model through our developer platform.

Open the platform

Get the model

Download the open weights and model card.

View on Hugging Face