G github.com Petals runs Llama 2 (70B) from Colab at 5 tokens/sec 5 points borzunov 3 years ago 3 comments
