neuralmagic
/

Llama-3.2-90B-Vision-Instruct-FP8-dynamic

Text Generation

compressed-tensors

Model card Files Files and versions Community

Resources

View closed (0)

Model does not run with VLLM

#3 opened 18 days ago by

Any idea when the evaluation data will be in for this model? would like to know how the performance differ from unquantized version of the model

#2 opened 2 months ago by

Any chance your team is working on a 4-bit Llama-3.2-90B-Vision-Instruct-quantized.w4a16 version?

#1 opened 3 months ago by