Up to four models
Llama 3.3 70B vs Llama 3.1-70B
On the index (GPT-4 = 100), Llama 3.3 70B scores 108 and Llama 3.1-70B 106. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
10892–124
WeightsOpen
ReleasedDec 2024
Price in / out$0.71 / $0.71
Context131K
Core tests taken1 of 9
10689–122
WeightsOpen
ReleasedJul 2024
Price in / out$0.72 / $0.72
Context131K
Core tests taken1 of 9
Llama 3.3 70B and Llama 3.1-70B are too close to call (2 points apart, within the uncertainty), ahead on 1 of 1 shared core tests, at a similar price.
Test by testthird-party results
Every result as a table
| Test | Llama 3.3 70B | Llama 3.1-70B |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 47.4% | 44.2% |
| MMLU (HELM runs) | 79.1% | 80.1% |
| Maths | ||
| AIME-style maths (OTIS mock) | 5.1% | 3.6% |
| MATH (level 5) | 41.6% | 36.7% |
| Coding | ||
| WeirdML | 14.4% | 9.0% |
| Agents | ||
| BALROG | 23.0% | 27.9% |
| DTBench | 59.5% | 60.0% |
| LMCA | 17.5% | 14.8% |
Tests at least two of these models have taken. Bold rows are the core tests.