Unverified50% confidenceBenchmarkExact time
In the archived Qwen experiments, the study conducted a gradient writing test on Qwen3.8-Flash-Next-FP8, and the specified first answer token was ranked highest among 132 independently reset targets, of which 93 were newly promoted and 39 were already in first place.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
VerifiedQwen3.8-Flash-Next achieved scores of 58.7 and 62.5 on the SWE-bench 1.1 and SWE-bench Pro evaluations, respectively.69% similarUnverifiedQwen3.8-27B scored 91.1 on OmniDocBench v1.5, the strongest document parsing score among the generative models in this evaluation.69% similarUnverifiedQwen3.8-Flash-Next has a 48-layer structure with 512 experts per layer and 180B-class parameters (125B main body + 51B N-gram table + 4B MTP).68% similarUnverifiedIn the tool-calling benchmark BFCL, Qwen3 scored 83%.66% similarUnverifiedOn the Qwen-3.5-4B-Instruct model, the top two principal axes matched the style dimensions of human spontaneous requests with a precision of 72.8% and a macro-averaged recall of 43.6%.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1046878API
curl https://kongchang.com/api/v1/knowledge/claims/1046878MCP
get_claim(id=1046878)