Unverified50% confidenceBenchmarkExact time
In a real-world test of 1,000 actual job posting data points, the decision model Jev was 4.8 times faster than the general-purpose large model Gemini
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/9/2026
First Seen
Valid until: 1/7/2027
Sources
Related Entities
Related Claims
Unverified移除歧义后所有模型准确率提升26.8至40.2个百分点，Gemini甚至达到81%，表明歧义是深度搜索代理最大性能瓶颈之一67% similarUnverifiedIn that test, the decision model Jev's accuracy trailed Gemini by only about 1.6 percentage points66% similarUnverifiedThe host's own testing found that minimal mode completes tasks approximately one-third faster than standard mode, with results comparable to standard mode.65% similarUnverifiedLiang Wenfeng estimates that training a top-tier large model on Huawei Ascend requires approximately 4 times the number of cards needed with NVIDIA.64% similarUnverifiedIn the fourth round of fusion experiments, treating JEV as the 9th framework and taking the highest score yielded an overall accuracy of 81.5%, but parity dropped back to 50%.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/992655API
curl https://kongchang.com/api/v1/knowledge/claims/992655MCP
get_claim(id=992655)