Llama 3-8B
ECI#203
109Llama 3-8B is an open-weights model from Meta AI, released April 18, 2024. It has an ECI score of 109 and ranks 203rd of 240 tracked models. At the time of its release, it ranked 17th of 50 tracked models.
Llama 3-8B vs select alternatives
| Domain | Benchmark | Llama 3-8B | |||
|---|---|---|---|---|---|
ECI scoreBest score 173 | 109 | 106 | 111 | 103 | |
| Ranking | #203 | #207 | #201 | #208 | |
| Conceptual reasoning | DTBenchBest score 99% | ||||
| Language understanding | Adversarial NLIBest score 58% | ||||
| Long context | ForecastBenchBest score 66.7 | 59.1 | 50.2 | — | — |
| Mathematics | MATH Level 5Best score 98% | ||||
OTIS Mock AIME 2024-2025Best score 100% | |||||
| Science | OpenBookQABest score 88% | ||||
ARC AI2Best score 95% | |||||
GPQA DiamondBest score 97% | |||||
| World knowledge | MMLUBest score 88% |
