Llama 3.1 70B Instruct

Meta llm

Meta's 70B-parameter multilingual instruction-tuned model from the Llama 3.1 family, positioned between the 8B and 405B variants.

overall ⌄

28.9

89 of 93 ranked

price · $/M tokens

self-hosted

open weights, no first-party API

to run it yourself ⌄

42 GB

minimum · 168 GB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmarkpublishedmeasured
MMLU-Pro66.4
GPQA48

Lineage

view in full graph →

Metadata

orgMeta
released2024-07-23
params70B
licensellama-3.1
hf idmeta-llama/Llama-3.1-70B-Instruct