Unverified50% confidenceBenchmarkExact time
On Autonomous Science Bench, SciMARS + Memory solved 12 of 18 GPT-6-Astra attempts, while the base multi-agent team solved only 8/18.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/10/2026
First Seen
Valid until: 11/24/2026
Sources
Related Entities
Related Claims
UnverifiedOn ScienceAgentBench, SciMARS + Memory solved 14 of 38 tasks, while the base team solved 11/38.72% similarUnverifiedIn the Artificial Analysis overall intelligence index, GPT-6 Sol scores a full 48 points (in the same tier as Astra), and Luna scores a full 36 points.68% similarUnverifiedAll Claude, GPT-6.1, and Gemini models achieved a perfect score of 12/12 in the experiment, with only 1 accidental impersonation response out of 252 attempts.68% 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 28.8.64% similarUnverifiedIn 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.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1154740API
curl https://kongchang.com/api/v1/knowledge/claims/1154740MCP
get_claim(id=1154740)