Unverified50% confidenceFactExact time
The llama.cpp inference framework does not natively support the FP8 format, and Q8_0 is not equivalent to FP8
1
Sources
50%
Confidence
Long-term
Relevance
8/6/2026
First Seen
Sources
Q8_K_XL与MXFP4命名之争：FP8≠Q8_0的量化真相
Related Claims
Unverifiedllama.cpp 支持从 Q2 到 Q8 的多种量化格式72% similarVerifiedThe llama.cpp project brought the GGUF quantization format into the mainstream, supporting multiple precision levels ranging from Q2_K to Q8_0.65% similarUnverified该Bug源于llama.cpp的fp16快速路径标志位未将sm_60（P100）排除，而sm_61（GTX 10系、P40）早已被排除63% similarUnverifiedFor users of Turing architecture GPUs, llama.cpp's GGUF quantization format is a better-compatibility alternative, as it does not depend on FP8 Tensor Core support.63% similarUnverifiedllama.cpp 用纯 C/C++ 重新实现了 LLaMA 模型的推理逻辑，无需 GPU 仅凭 CPU 就能运行大语言模型60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/697308API
curl https://kongchang.com/api/v1/knowledge/claims/697308MCP
get_claim(id=697308)