Search: “berkeley-ai-research”

2 results2 results

Model Releases

From CUDA to MLX: How K-Search Brings Decades of Kernel Expertise to Apple Silicon

DGX agent

Figure 1: CUDA-to-MLX optimization translation map. CUDA optimization knowledge can be translated into architecture-native MLX strategies rather than copied instruction-for-instruction. We face a new

model-releasesberkeley-ai-research

29 Jul 2026

Model Releases

Teaching LLMs to Update Beliefs for Efficient Long-Horizon Interaction

DGX agent

--> .abbel-fig { display: block; text-align: center; margin: 2.4em 0; line-height: 1.4; max-width: 100%; } .abbel-fig img { display: block; margin: 0.65em auto 0; height: auto; max-width: 100%; } /* I

model-releasesberkeley-ai-research

26 Jul 2026

2 results