Tencent releases NVFP4-quantized WeMM-Video-9B checkpoint
Tencent published an NVFP4 W4A16 Safetensors variant of WeMM-Video-9B, with 13.07 GiB of weight files. The checkpoint accepts text, images, audio and video and generates Chinese or English responses, according to Tencent's Hugging Face repository; deployment uses patched vLLM 0.31.0 with ModelOpt Mixed Precision and Marlin.
This adds a lower-precision deployment option for Tencent's 9B omni-modal model while retaining joint audio, visual and text inputs.
tencent/WeMM-Video-9B-NVFP4 Video-Text-to-Text • 10B • Updated about 4 hours ago • 5
license: apache-2.0 language:
- en
- zh pipeline_tag: video-text-to-text base_model: tencent/WeMM-Video-9B base_model_relation: quantized tags:
- wemm
- multimodal
- video
- audio
- quantized
- nvfp4
WeMM-Video-9B · NVFP4 W4A16
Overview
This repository contains the NVFP4 W4A16 Safetensors variant of WeMM-Video-9B. WeMM-Video is a Chinese–English bilingual family for joint audio-visual understanding and reasoning. It accepts text, images, audio, and video and generates text responses.
#Tencent#WeMM-Video-9B-NVFP4#WeMM-Video#vLLM#ModelOpt#Marlin