BENCHMARK REGISTRY
Public Artificial Intelligence Benchmark Records Database

Llama 3.1 Instruct 70B

MODEL RECORD: META-LLAMA31INSTRUCT70B

Organization
Meta
Family
LLAMA31INSTRUCT70B
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-009LLAMA31INSTRUCT70B-001 Llama 3.1 Instruct 70B Artificial Analysis Intelligence Index Overall 6.8 Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-002 Llama 3.1 Instruct 70B GPQA Diamond Accuracy 40.9% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-003 Llama 3.1 Instruct 70B Humanity's Last Exam No tools Overall 4.6% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-004 Llama 3.1 Instruct 70B SciCode Overall 26.7% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-005 Llama 3.1 Instruct 70B tau2-bench Overall 15.2% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-006 Llama 3.1 Instruct 70B Terminal-Bench Hard Overall 3% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-007 Llama 3.1 Instruct 70B CritPt Overall 0% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-008 Llama 3.1 Instruct 70B AA-Omniscience Overall -40.8 Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-009 Llama 3.1 Instruct 70B AA-Omniscience Accuracy 19.4% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-010 Llama 3.1 Instruct 70B IFBench Overall 34.4% Unknown Primary source META-LLAMA31INSTRUCT70B
BR-009LLAMA31INSTRUCT70B-011 Llama 3.1 Instruct 70B AA-LCR Overall 6.3% Unknown Primary source META-LLAMA31INSTRUCT70B