Meta
Llama 3-70B benchmark results
6 results on 6 benchmarks from 3 sources.
Open weights Released: Apr 18, 2024 Updated Oct 22, 2026
Also listed as: Meta-Llama-3-70B-Instruct · Llama 3 (70B) · Llama-3-70B · meta-llama/Meta-Llama-3-70B-Instruct · meta-llama/Meta-Llama-3-70B
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| Open LLM Leaderboard v2 | 36.4% | 534 | Average of 6 normalised benchmarks; bfloat16; 70.554B params“meta-llama/Meta-Llama-3-70B-Instruct” | Official leaderboard | Mar 13, 2025 |
| GPQA Diamond | 40.6% | 238 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| MMLU | 79.3% | 21 | Exact match; 5-shot | Official leaderboard via Epoch AI | Apr 18, 2024 |
| MMLU-Pro | 52.8% | 158 | Overall accuracy; data source: TIGER-Lab; 70B“Llama-3-70B” | Official leaderboard | Mar 11, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 4.3% | 237 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Feb 25, 2025 |
| MATH Level 5 | 22.6% | 86 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
FAQ
Where do these Llama 3-70B scores come from?
From Epoch AI Benchmarking Hub, MMLU-Pro leaderboard (TIGER-Lab), and Hugging Face Open LLM Leaderboard v2 (archived). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.