dvmazur/mixtral-offloading · AI Repo StatsDVdvmazur/mixtral-offloading
Run Mixtral-8x7B models in Colab or consumer desktops
Training & fine-tuningPython MIT 224created 2023-12-15 #colab-notebook#deep-learning#google-colab#language-model#llm#mixture-of-experts#offloading#pytorch#quantization