Tell me how do you feel about this model without telling me how do you feel about this model
I'll start...
111B is quite a good size, actually. It possible to run a good quant with decent context length on just four 24GB consumer-grade GPUs. And it is even a bit smaller than Mistral Large 123B and Pixtral Large 124B, which I use daily.
111B is quite a good size, actually. It possible to run a good quant with decent context length on just four 24GB consumer-grade GPUs. And it is even a bit smaller than Mistral Large 123B and Pixtral Large 124B, which I use daily.
Haha, yeah let me just grab the spare four 24GB consumer-grade GPUs from my garage, they would collect the dust otherwise... 🤣
C4AI Command A is an open weights research release
is cc-by-nc
Calling yet another cc-by-nc (except for the for-profit parent org) release “open weights” is a pretty good joke by co$here
C4AI Command A is an open weights research release
is cc-by-nc
Calling yet another cc-by-nc (except for the for-profit parent org) release “open weights” is a pretty good joke by co$here
"open weights research" whats your issue