Tulu 3 (Tülu 3) 70B
| Released | 2024-11-21 |
|---|---|
| Price per 1M | Not listed |
| Overall score | 35.1 (early, few results) |
| Provider | Ai2 |
| Weights | Open weights |
Scores
| Dimension | Score | Rank | Benchmarks |
|---|---|---|---|
| Overall: Balanced | 35.1 | 3 | |
| Overall: Coding | 36.1 | 1 | |
| Overall: Reasoning | 35.0 | 3 | |
| Overall: Agentic work | 36.1 | 1 | |
| Reasoning | 36.1 | 1 | |
| Math | 33.7 | 2 |
Evidence
Reasoning 36.1 ±4.6
| Benchmark | Result | Source | Date |
|---|---|---|---|
| GPQA Diamond | 46.3% | Epoch AI | 2025-01-27 |
Math 33.7 ±2.0
| Benchmark | Result | Source | Date |
|---|---|---|---|
| MATH Level 5 | 42.7% | Epoch AI | 2025-01-27 |
| OTIS Mock AIME 2024-2025 | 4.4% | Epoch AI | 2025-03-07 |
Names matched to this model
llama-3-1-tulu-3-70b-dpo, tulu-3-t-lu-3-70b