Unverified50% confidenceBenchmarkExact time
Qwen3.5-4B achieved a 10.11 percentage point improvement in accuracy on OSWorld-G.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedQwen3 27B's score on OS World Verified jumped from 63.9 to 84.3 (self-reported official data).70% similarUnverifiedThe fine-tuned Qwen3.5-4B model improved accuracy on BU Bench V1 from 15% to 53%.68% similarUnverifiedQwen 3.6 Max has an end-to-end accuracy of only 12.3% in neutral mode68% similarUnverifiedVia --quantization quark_mxfp4, SGLang dequantizes NVFP4 weights and requantizes them to MXFP4 at load time, achieving GSM8K accuracy recovery of 97.5%–100.2% on models such as MiniMax-M2.7 and GLM-5.1.65% similarUnverifiedOsprey achieves a 16.1% improvement in mean acceptance length on Qwen3-8B.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975838API
curl https://kongchang.com/api/v1/knowledge/claims/975838MCP
get_claim(id=975838)