Introducing Mellum2: A 12B Mixture-of-Experts Model by JetBrains

Accelerate LLM model loading and increase context windows with GPUDirect on Amazon FSx for Lustre and TurboQuant | Artificial Intelligence