13 points | by marcobambini 2 days ago ago
4 comments
Read the readme but still unclear - does this mean we can run DeepSeek v4.1 Flash on GPUs with 8GB VRAM?
looks promising, will test it out.
What are the usecase that you can think of because 3.77 tok/s is actually to slow and better call a cloud API to do the task
Any well defined problem would probably be of use. How often anyone comes across these without other help is anyones guess.
Read the readme but still unclear - does this mean we can run DeepSeek v4.1 Flash on GPUs with 8GB VRAM?
looks promising, will test it out.
What are the usecase that you can think of because 3.77 tok/s is actually to slow and better call a cloud API to do the task
Any well defined problem would probably be of use. How often anyone comes across these without other help is anyones guess.