BENCHMARK REGISTRY
Public Artificial Intelligence Benchmark Records Database

Claude 3.5 Haiku

MODEL RECORD: ANTHR-H35

Organization
Anthropic
Family
Claude
Model number
35
Tier
H
Status
ACTIVE
Release date
2024-10-22

Table may be scrolled horizontally on narrow displays.

Record No. Model Benchmark Metric Score Eval Date Source Model ID
BR-002H35-001 Claude 3.5 Haiku Artificial Analysis Intelligence Index Overall 12.3 Unknown Primary source ANTHR-H35
BR-002H35-002 Claude 3.5 Haiku GPQA Diamond Accuracy 40.8% Unknown Primary source ANTHR-H35
BR-002H35-003 Claude 3.5 Haiku Humanity's Last Exam No tools Overall 3.5% Unknown Primary source ANTHR-H35
BR-002H35-004 Claude 3.5 Haiku SciCode Overall 27.4% Unknown Primary source ANTHR-H35
BR-002H35-005 Claude 3.5 Haiku tau2-bench Overall 24.6% Unknown Primary source ANTHR-H35
BR-002H35-006 Claude 3.5 Haiku tau3-Banking Overall 5.8% Unknown Primary source ANTHR-H35
BR-002H35-007 Claude 3.5 Haiku Terminal-Bench Hard Overall 2.3% Unknown Primary source ANTHR-H35
BR-002H35-008 Claude 3.5 Haiku Terminal-Bench 2.1 Overall 10.1% Unknown Primary source ANTHR-H35
BR-002H35-009 Claude 3.5 Haiku CritPt Overall 0% Unknown Primary source ANTHR-H35
BR-002H35-010 Claude 3.5 Haiku AA-Omniscience Overall -23.2 Unknown Primary source ANTHR-H35
BR-002H35-011 Claude 3.5 Haiku AA-Omniscience Accuracy 13% Unknown Primary source ANTHR-H35
BR-002H35-012 Claude 3.5 Haiku IFBench Overall 42.8% Unknown Primary source ANTHR-H35
BR-002H35-013 Claude 3.5 Haiku GDPval-AA Average Normalized Score 0% Unknown Primary source ANTHR-H35
BR-002H35-014 Claude 3.5 Haiku MMMU-Pro Overall 45.6% Unknown Primary source ANTHR-H35
BR-002H35-015 Claude 3.5 Haiku AA-LCR Overall 23.3% Unknown Primary source ANTHR-H35