Unverified50% confidenceBenchmarkExact time
In the GSM8K test (1319 questions, 8192-token limit), BF16 scored 1249, FP8 scored 1241, MXFP8 scored 1250, and NVFP4 scored 1238, with the three quantized formats matching BF16 within a ±3 point tolerance.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
UnverifiedIn this test, when reading a 2048-token prompt, FP8 prefill speed was 10,118 tok/s, ahead of BF16 (7,003 tok/s), NVFP4 (6,978 tok/s), MXFP8 (6,061 tok/s) and GGUF Q5_K_M (1,710 tok/s).73% similarUnverifiedBF16 uses 1 sign bit, 8 exponent bits, and 7 mantissa bits, sacrificing precision while retaining the same dynamic range as FP32.70% similarUnverifiedIn the token-by-token fidelity test, NVFP4 had an 89%–90% probability of choosing the same next token as BF16, while FP8 and MXFP8 reached 94%–96%.68% similarUnverifiedIn GGUF, BF16 uses 16 bits per parameter, Q8_0 approximately 8 bits, Q4_K_M approximately 4 bits, Q3_K_M approximately 3 bits, and Q2_K approximately 2 bits.67% similarUnverifiedIn the task of completing the last Token of an FP8 quantization function, after DeepSeek-V4-Flash-Base adds equal-sign decorations before the code, when the number of equal signs modulo 4 is 0 or 1 the error rate is as high as 71.3%, and when the remainder is 2 or 3 the accuracy returns to 91.5%.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1057660API
curl https://kongchang.com/api/v1/knowledge/claims/1057660MCP
get_claim(id=1057660)