deepseek-ai

/

deepseek-coder-33b-instruct

like 454

Follow

DeepSeek 811

working on this

thanks yeah, I saw that, I wanted the tokenizer.model to make ggufs of fine-tunes. btw, llama.cpp is going to supporting quanting without the tokenizer.model so we're all set

GPTQ exllama also needs that tokenizer.model as well