canada-quant
/

DeepSeek-V4-Flash-W4A16-FP8

Text Generation

compressed-tensors

mixture-of-experts

Mixture of Experts

Model card Files Files and versions

Resources

View closed (1)

Small patch to fix MTP on current vLLM (8th of June, 26)

#8 opened 4 months ago by

2026-05-26 dual-spark bench: GSM8K 94-98%, AIME 4/5, concurrency wall 12→125 tok/s

#7 opened 5 months ago by

Compatible version with Ampere? SM8.6

#6 opened 5 months ago by

One shot bootstrap for 2 Sparks not working

#4 opened 5 months ago by

dgx spark error ：torch.AcceleratorError: CUDA error: an illegal memory access was encountered ......

#3 opened 5 months ago by

Need vllm-w4a16-dsv4:exp Thanks

#2 opened 5 months ago by

Can I run this model on 2x H20 141GB?

#1 opened 5 months ago by