INT21 vs Rudrite Research

We are building self-improving AI systems for the software beneath modern AI. Our first product generates and optimizes low-level GPU software, then proves its work with...

Interactive, animated, visual explainers of landmark AI & ML papers, attention & transformers, GPT-3, FlashAttention, Mixtral, Mamba, DeepSeek, and more.

About INT21

We are building self-improving AI systems for the software beneath modern AI. Our first product generates and optimizes low-level GPU software, then proves its work with tests and benchmarks. Today, we are launching INT21: the first company to achieve self-improving AI agent swarms, applied to AI infrastructure. That software has an outsized effect on speed and cost. The first four implementations produced by PTX Kernel Factory are open source today. The product is also entering beta, with early access available at int21.ai.

About Rudrite Research

Interactive, animated, visual explainers of landmark AI & ML papers, attention & transformers, GPT-3, FlashAttention, Mixtral, Mamba, DeepSeek, and more. Interactive, animated, visual explainers of landmark AI & ML papers, the systems and ideas behind the models you use, redrawn and made legible. Direct Preference Optimization: Your Language Model is Secretly a Reward Model. Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale.