Unverified50% confidenceBenchmarkExact time
GPT-6 LUNA MAX's average score on visual grounding tests improved from 28.8% before the fix to 60% after the fix.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/8/2026
First Seen
Valid until: 1/6/2027
Sources
Related Entities
Related Claims
UnverifiedIn EvoArena testing, GPT-5.5, which achieved the highest Step Level score, had an accuracy of just over 50%, while Gemini-3.1 Pro and GLM-5.1, which performed best at the Chain Level, reached only approximately 28%–29%.71% similarUnverifiedOn Artificial Analysis's Omniscience hallucination metric, GPT-5.6 Sol Max reached 89%, higher than Claude Fable's 55% and GLM 5.2's 28%.71% similarVerifiedIn the Agents' Last Exam benchmark, GPT-5.6 Sol achieved a score of 53.6, which is 13.1 points higher than Claude Fable 5 using adaptive reasoning.71% 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.69% similarUnverified在长周期智能体工作流测试（Engines Last Exam）中，GPT-5.6 Sol 得分 53.6，比 Faber-5 高出 13.1 分69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/988754API
curl https://kongchang.com/api/v1/knowledge/claims/988754MCP
get_claim(id=988754)