Codesota · Models · Llama 3.1 70BMeta4 results · 4 benchmarks
Model card

Llama 3.1 70B.

Metaopen-sourceLLM

Meta's 70B parameter open-source model.

§ 02 · Benchmarks

Every benchmark Llama 3.1 70B has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01HumanEvalCode & Software Engineering · Code Generationpass@180.5%#30/332026-04-20source ↗
02MATHLanguage & Knowledge · Mathematical Reasoningaccuracy68.0%#27/292026-04-20unverified
03MMLU—accuracy82.0%#17/192026-04-20unverified
04GPQA—accuracy41.7%#16/172026-04-20unverified
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where Llama 3.1 70B actually performs.

Language & Knowledge
1
benchmark
avg rank #27.0
Code & Software Engineering
1
benchmark
avg rank #30.0
§ 06 · Sources & freshness

Where these numbers come from.

src
3
results
codesota-api
1
result
0 of 4 rows marked verified.