Meta
Llama 3-8B benchmark results
8 results on 6 benchmarks from 3 sources.
Open weights Released: Apr 18, 2024 Updated Oct 22, 2026
Also listed as: Meta-Llama-3-8B-Instruct · Llama 3 (8B) · Llama-3-In 8b · Llama-3-8B · meta-llama/Meta-Llama-3-8B-Instruct · meta-llama/Meta-Llama-3-8B
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| Open LLM Leaderboard v2 | 23.9% | 1868 | Average of 6 normalised benchmarks; bfloat16; 8.03B params“meta-llama/Meta-Llama-3-8B-Instruct” | Official leaderboard | Mar 13, 2025 |
| Open LLM Leaderboard v2 | 20.6% | 2570 | Average of 6 normalised benchmarks; float16; 8.03B params“meta-llama/Meta-Llama-3-8B-Instruct” | Official leaderboard | Mar 13, 2025 |
| GPQA Diamond | 26.1% | 272 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
| MMLU | 66.5% | 37 | Exact match; 5-shot | Paper / tech report via Epoch AI | Apr 18, 2024 |
| MMLU | 68.8% | 58 | Exact match; 5-shot | Official leaderboard via Epoch AI | Apr 18, 2024 |
| MMLU-Pro | 35.4% | 205 | Overall accuracy; data source: TIGER-Lab; 8B“Llama-3-8B” | Official leaderboard | Mar 11, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 1.9% | 249 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 30, 2026 |
| MATH Level 5 | 6.1% | 101 | Epoch AI run, accuracy | Independent run | Jan 27, 2025 |
FAQ
Where do these Llama 3-8B 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.