𝚐𝔪𝟾𝚡𝚡𝟾 on Warpcast

Content pfp

0 reply

0 recast

0 reaction

𝚐𝔪𝟾𝚡𝚡𝟾 pfp

𝚐𝔪𝟾𝚡𝚡𝟾

codellama 70b is finally here. base: https://huggingface.co/codellama/CodeLlama-70b-hf

3 replies

1 recast

14 reactions

𝚐𝔪𝟾𝚡𝚡𝟾 pfp

𝚐𝔪𝟾𝚡𝚡𝟾

quantized CodeLlama 7b Python to 4-bit with MLX, -this model is now optimized for high-speed performance on apple silicon. https://huggingface.co/mlx-community/CodeLlama-7b-Python-4bit-MLX

0 reply

0 recast

5 reactions