Unverified50% confidenceFactExact time
The page linked to by the result is labeled with version information for tier-4-v2, indicating that the benchmark itself has undergone iterations and that directly comparing scores across versions is not rigorous.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/2/2026
First Seen
Valid until: 12/31/2026
Sources
Related Entities
Related Claims
UnverifiedIn terms of technical score, Step-5-preview ranked second.59% similarUnverifiedOn Qwen-3.5-4B-Instruct, 75.6% of validity scores judged that the generated results at both poles of the principal axes accurately corresponded to their labels, with an annotator adjacent-agreement rate as high as 90.9%.58% similarUnverified0.8B parameters places it in the same ultra-small model tier as Meta's LLaMA 3.2 1B and Google's Gemma 2B, close to the lower bound of early GPT-2 (1.5B) in scale.57% similarUnverifiedSpecific performance metrics for v2.5, quantitative comparisons with older versions, and details such as migration costs have not yet been made public.56% similarVerifiedQwen3-4B is already on par with DeepSeek V3 or GPT-4o on multiple reasoning evaluation metrics.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/968625API
curl https://kongchang.com/api/v1/knowledge/claims/968625MCP
get_claim(id=968625)