nvidia
/

DeepSeek-V4-Flash-NVFP4

Text Generation

Model Optimizer

8-bit precision

Model card Files Files and versions

DeepSeek-V4-Flash-NVFP4

Commit History

Fix benchmark note: remove inaccurate FP8 baseline line, correct max tokens to 384000

48bfe38
verified

jingyux-nv commited on about 8 hours ago

Remove Artificial Analysis (AA) reference numbers from evaluation table

56fbbe4
verified

jingyux-nv commited on about 16 hours ago

Update model card: add Flash evaluation results and vLLM/SGLang deployment sections

9ec8127
verified

jingyux-nv commited on 2 days ago

Keep MTP block in source precision

6ee35c6
verified

jingyux-nv commited on 3 days ago

Add files using upload-large-folder tool

163f205
verified

jingyux-nv commited on 20 days ago

Add files using upload-large-folder tool

6538918
verified

jingyux-nv commited on 20 days ago

initial commit

4f66995
verified

jingyux-nv commited on 20 days ago