BENCHMARK REGISTRY
Public Artificial Intelligence Benchmark Records Database

Llama 3.1 Instruct 405B

MODEL RECORD: META-LLAMA31INSTRUCT405B

Organization
Meta
Family
LLAMA31INSTRUCT405B
Model number
Unknown
Tier
Unknown
Status
ACTIVE
Release date
2024-07-23

Table may be scrolled horizontally on narrow displays.

Record No. Model Benchmark Metric Score Eval Date Source Model ID
BR-009LLAMA31INSTRUCT405B-001 Llama 3.1 Instruct 405B Artificial Analysis Intelligence Index Overall 8.5 Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-002 Llama 3.1 Instruct 405B GPQA Diamond Accuracy 51.5% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-003 Llama 3.1 Instruct 405B Humanity's Last Exam No tools Overall 4.2% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-004 Llama 3.1 Instruct 405B SciCode Overall 29.9% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-005 Llama 3.1 Instruct 405B tau2-bench Overall 19% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-006 Llama 3.1 Instruct 405B Terminal-Bench Hard Overall 6.8% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-007 Llama 3.1 Instruct 405B CritPt Overall 0% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-008 Llama 3.1 Instruct 405B AA-Omniscience Overall -17.3 Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-009 Llama 3.1 Instruct 405B AA-Omniscience Accuracy 22.3% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-010 Llama 3.1 Instruct 405B IFBench Overall 39% Unknown Primary source META-LLAMA31INSTRUCT405B
BR-009LLAMA31INSTRUCT405B-011 Llama 3.1 Instruct 405B AA-LCR Overall 24.3% Unknown Primary source META-LLAMA31INSTRUCT405B