Rankings / Allen Institute for AI,University of Washington

Llama 3.1 Tulu 3 70B Dpo

released 2024-11-21

BenchAtlas Index

as of 2026-07-14

Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

3 results

knowledge science

GPQA Diamond

46.3 %

independent

Epoch AI Benchmarking Hub

2025-01-27

reasoning math

MATH Level 5		42.7 %	independent	Epoch AI Benchmarking Hub 2025-01-27
OTIS Mock AIME 2024–2025		4.4 %	independent	Epoch AI Benchmarking Hub 2025-03-07