ModelCap
Current AI model comparison
Exact identities · public evidence
Meituan Longcat
LongCat Flash Thinking 2601
Position
#50
Index
60.1
Estimated
global-corpus-prior over 186 held-out anchors (12%); launch card against 4 resolved peers on 3 rows (88%); exceeds every named peer on 1 of 3 rows; optimism haircut 0.15 from cross-lab-probe; shrunk 1.1 toward the measured corpus
Comparison side 1
OpenAI
GPT-5.5 Pro
Position
#53
Index
58.2
Measured
publisher-corpus-prior over 186 held-out anchors (61%); 1 specialist-board observation at 2.1% support (39%); shrunk 14.5 toward the measured corpus
Comparison side 2
Output / 1M
Unavailable
vs $180
Context
—
vs 1M
Weights
Open weights
vs API only