Unverified50% confidenceBenchmarkExact time
On 194 Yelp text judgments, B1 (0.8B) got 184 correct and B2 (4B) got 192 correct; the average NLLs were 0.09620 and 0.04024, respectively.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
UnverifiedOn 1,184 historical multi-task questions, B1 got 951 correct and B2 got 1,033 correct, with NLLs of 0.60133 and 0.46749, respectively.77% similarUnverifiedIn 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.66% similarUnverifiedUnder standard DDPM settings, rank 4 achieved the best FID score of 124.1380, with rank 8 following closely at 124.2136.66% similarUnverifiedScores across 19 models ranged from 0.55 to 1.00, with Qwen 3 235B A22B Instruct ranking first at 1.00.66% similarUnverifiedThe self-developed W4 kernel's perplexity (probe ppl) is 15.76, better than bitsandbytes NF4's 18.98.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1064019API
curl https://kongchang.com/api/v1/knowledge/claims/1064019MCP
get_claim(id=1064019)