Unverified50% confidenceFactExact time
NVIDIA provides native support for nvfp4 in its TensorRT-LLM framework.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedNVIDIA recently introduced Multi-Device Inference Support for TensorRT75% similarUnverifiedNVIDIA在Hugging Face发布Qwen3 235B的NVFP4量化模型，NVFP4是针对Hopper和Blackwell架构的4-bit浮点格式，理论压缩比达4倍74% similarUnverifiedllmfit 通过调用 NVML 获取 NVIDIA GPU 的显存总量、可用量和架构型号，NVML 是 nvidia-smi 背后的同一套 C 语言 API 库74% similarUnverifiedNVIDIA designed a dedicated Tensor Core instruction mma.kind::mxf4 for NVFP4 in the Blackwell architecture (SM 100/103, such as the B100 and B200 series), and adopted an FP8 scaling factor mechanism with one scale per group of 16 elements.74% similarUnverifiedNVIDIA NIM内置了TensorRT-LLM等推理加速引擎74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/894790API
curl https://kongchang.com/api/v1/knowledge/claims/894790MCP
get_claim(id=894790)