Unverified50% confidenceFactExact time
The V4.1 architecture is robust to quantization, with accuracy loss on the MMLU benchmark of less than 0.2% when switching from FP16 to FP8.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedThe M5 Ultra's FP8 and FP4 floating-point compute performance is identical to its BF16 performance, and the gap with the Spark narrows noticeably when dropping to FP4 precision.76% similarUnverifiedV100的Tensor Core在FP16精度下峰值算力可达125 TFLOPS，是前代Pascal架构的5倍以上73% similarUnverifiedTensorRT supports multiple precision calibrations including FP32, FP16, INT8, and FP871% similarUnverifiedNVFP4 quantization is not 4-bit for the whole model: the vision encoder, connector, Mamba convolutions, and embedding stay in BF16, and the MoE router stays in FP32.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1067169API
curl https://kongchang.com/api/v1/knowledge/claims/1067169MCP
get_claim(id=1067169)