ARC-AGIラボによる報告
93.7%inferred family alias from gpt-5.4 (score=0.3050
BrowseCompラボによる報告
82.7%inferred family alias from gpt-5.4 (score=0.3050
54.6%inferred family alias from gpt-5.4 (score=0.3050
56%inferred family alias from gpt-5.4 (score=0.3050
FrontierMathラボによる報告
47.6%inferred family alias from gpt-5.4 (score=0.3050
92.8%inferred family alias from gpt-5.4 (score=0.3050
Graphwalks BFS >128kラボによる報告
21.4%inferred family alias from gpt-5.4 (score=0.3050
Graphwalks parents >128kラボによる報告
32.4%inferred family alias from gpt-5.4 (score=0.3050
73.3%inferred family alias from gpt-5.4 (score=0.3050
Graphwalks bfs <128kラボによる報告
93%inferred family alias from gpt-5.4 (score=0.3050
Graphwalks parents <128kラボによる報告
89.8%inferred family alias from gpt-5.4 (score=0.3050
39.8%inferred family alias from gpt-5.4 (score=0.3050
MMMU Proラボによる報告
81.2%inferred family alias from gpt-5.4 (score=0.3050
OmniDocBench 1.5ラボによる報告
89.1%inferred family alias from gpt-5.4 (score=0.3050
75%inferred family alias from gpt-5.4 (score=0.3050
Scale MCP Atlasラボによる報告
67.2%inferred family alias from gpt-5.4 (score=0.3050
SWE Bench Proラボによる報告
57.7%inferred family alias from gpt-5.4 (score=0.3050
Tau 2 Telecomラボによる報告
98.9%inferred family alias from gpt-5.4 (score=0.3050
Terminal Bench 2.0ラボによる報告
75.1%inferred family alias from gpt-5.4 (score=0.3050