BENCHMARK REGISTRY
Public Artificial Intelligence Benchmark Records Database

Inception

Provider prefix
INCPT
Benchmark Record namespace
041

Table may be scrolled horizontally on narrow displays.

Record No. Model Benchmark Metric Score Eval Date Source Model ID
BR-041MERCURY2-001 Mercury 2 Artificial Analysis Intelligence Index Overall 25.3 Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-002 Mercury 2 GPQA Diamond Accuracy 77% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-003 Mercury 2 Humanity's Last Exam No tools Overall 15.5% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-004 Mercury 2 SciCode Overall 38.7% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-005 Mercury 2 tau2-bench Overall 70.8% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-006 Mercury 2 Terminal-Bench Hard Overall 26.5% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-007 Mercury 2 CritPt Overall 0.8% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-008 Mercury 2 AA-Omniscience Overall -52.3 Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-009 Mercury 2 AA-Omniscience Accuracy 20.5% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-010 Mercury 2 IFBench Overall 69.8% Unknown Primary source INCPT-MERCURY2
BR-041MERCURY2-011 Mercury 2 AA-LCR Overall 36.3% Unknown Primary source INCPT-MERCURY2