Verified90% confidenceFactTime unknown
BF16 (Brain Floating Point 16) is a floating-point variant proposed by Google that retains the 8-bit exponent range of FP32 but reduces the mantissa to 7 bits.
7
Sources
90%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedBF16采用1位符号位、8位指数位和7位尾数位，牺牲精度但保留了与FP32相同的动态范围73% similarVerifiedCompared to FP16, BF16 has the same 8-bit exponent width as FP32 and therefore a larger dynamic range, making it the mainstream precision choice for current large model training.65% similarUnverifiedThe conversion from FP8 to BF16 is lossless because BF16 can exactly represent all values representable by FP8 E4M3, while the reverse is not true61% similarVerifiedFP32每个参数占4字节，FP16/BF16占2字节，INT8量化占1字节，INT4量化占0.5字节60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/21171API
curl https://kongchang.com/api/v1/knowledge/claims/21171MCP
get_claim(id=21171)