Unverified50% confidenceFactExact time
The experiment wrote a total of 2,783 programs, died 605 times, and ran continuously for 27 hours.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/17/2026
First Seen
Valid until: 1/15/2027
Sources
Related Entities
Related Claims
UnverifiedThe system retrieves news related to a question from the past 60 days, filters out content referencing odds, retains the 8 most relevant articles, produces a total of 27 decisions per question, and completes the entire process in approximately 2 seconds.64% similarUnverifiedThe EgoDex dataset contains 338K episodes, 194 task types, and 829 hours of recorded footage63% similarUnverifiedCompleting the 30-second version involved 14 sub-agents working in coordination, with the entire process involving 779 tool calls (161 in the main conversation and 618 in sub-agents), taking approximately 2.4 hours in total.63% similarUnverifiedThe entire modeling process consumed approximately 17 million input tokens and 275,000 output tokens, with a total duration of approximately 1 hour and 8 minutes.62% similarVerifiedOSWorld 2.0 contains 108 long-workflow real computer tasks, with a median completion time of 1.6 hours for skilled humans, approximately 48 times that of version 1.0 (2 minutes).61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1049805API
curl https://kongchang.com/api/v1/knowledge/claims/1049805MCP
get_claim(id=1049805)