Meta AI·released 2024-04-181 source
Llama 3-70B benchmark scores: 6 benchmarks tracked. 50% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 6 benchmarks·best result 72.4 on MMLU (unverified)·3 of 6 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| MMLU | 72.4accuracy (%) | unverified· optimizedT1 | 2024-04-18Stanford CRFM Leaderboard |
| Winogrande | 67accuracy (%) | unverified· optimizedT1 | 2024-04-18The Llama 3 Herd of Models |
| MATH level 5 | 22.55accuracy (%) | reproduced· optimizedT1 | 2024-04-18Epoch AI |
| GPQA diamond | 20.75accuracy (%) | reproduced· optimizedT1 | 2024-04-18Epoch AI |
| Cybench | 5accuracy (%) | unverified· optimizedT1 | 2024-04-18Cybench leaderboard |
| OTIS Mock AIME 2024-2025 | 4.21accuracy (%) | reproduced· optimizedT1 | 2024-04-18Epoch AI |
Claims drawn from cited facts, not live model generation.
This model was developed by Meta AI and released in 2024.
2 cited facts
What stands out is the verification: these 6 scores are independently reproduced, so even a lead-free record is solid ground — measured results, not vendor claims.
3 cited facts
This model trails the state-of-the-art leader by an average of 59.09 points, which is a wide gap indicating it is well behind the leaders.
1 cited fact
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
Llama 3-70B is an AI model developed by Meta AI, released 2024-04-18. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Llama 3-70B has recorded scores on 6 benchmarks, each shown with its evidence status.
Llama 3-70B has recorded scores on 6 benchmarks — Winogrande, GPQA diamond, MATH level 5, OTIS Mock AIME 2024-2025, MMLU, Cybench. The full table above shows each score with its evidence status.
3 of 6 recorded scores (50%) are independently reproduced rather than self-reported by the lab.
Llama 3-70B has 6 tracked claims: 3 independently reproduced, 3 unverified.