Up to four models
Claude 3 Haiku vs Ministral 3B
On the index (GPT-4 = 100), Claude 3 Haiku scores 95 and Ministral 3B 93. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude 3 Haiku and Ministral 3B are too close to call (2 points apart, within the uncertainty), ahead on 1 of 1 shared core tests.
Test by testthird-party results
Claude 3 Haiku
Ministral 3B
Every result as a table
| Test | Claude 3 Haiku | Ministral 3B |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 36.3% | 25.3% |
| Maths | ||
| MATH (level 5) | 14.9% | 14.4% |
| Agents | ||
| DTBench | 50.1% | 51.7% |
| LMCA | 8.8% | 5.5% |
Tests at least two of these models have taken. Bold rows are the core tests.