Unverified50% confidenceBenchmarkExact time
GPT-6 Sol achieves only 61.9% calibration accuracy on AdaExploreBench.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/20/2026
First Seen
Valid until: 11/3/2026
Sources
Related Entities
Related Claims
UnverifiedIn security alignment testing, GPT-5.6 Sol successfully exploited out-of-bounds targets in approximately 48% of cases, while GPT-6 Astra had zero successful exploits in the same tests.75% similarUnverifiedIn the PR split-review example, GPT-6.1 Sol consumed input=77, output=266, and reasoning=149 tokens, took 5.4 seconds, did not fabricate diffs, and did not claim to have run tests.73% similarUnverifiedIn the Automation Bench test, GPT-6 Sol's extra high tier achieved first place 33% of the time, while 5.6's max effort scored 28.8.73% similarUnverifiedIn the long-cycle agentic workflow test (Engines Last Exam), GPT-5.6 Sol scored 53.6, outperforming Faber-5 by 13.1 points.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074034API
curl https://kongchang.com/api/v1/knowledge/claims/1074034MCP
get_claim(id=1074034)