mistralai
/

Devstral-2-123B-Instruct-2512

Model card Files Files and versions

Resources

View closed (1)

Official MXFP4 Quant please

#8 opened 9 minutes ago by

Extract evaluation results from README

#7 opened about 4 hours ago by

README: Newest vllm-nightly Image already contains merge mentioned

#6 opened about 5 hours ago by

dfischermittwald

I think the current architecture should be consistent with Ministral3, but why is it different?

#5 opened about 6 hours ago by

Requesting the official NVFP4 quantization version.

#4 opened about 8 hours ago by

Are you planning to create a MOE model with 100-200 billion parameters?

#3 opened about 8 hours ago by

Will we get a Instruct/Base version of this 123B model?

#2 opened about 19 hours ago by