Claude 3 Sonnet is the stronger all-rounder.
Scores updated · 16 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Coding rests on a single test.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Claude 3 Sonnet pulls ahead
- Code for real scientific research problemsSciCode+11.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+6.9points ahead
Where DBRX Instruct pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 16 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningClaude 3 Sonnet
- GPQA DiamondClaude 3 Sonnet by 6.94033.1+6.9
- Humanity's Last Examtie3.62.9tie
Other results13 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- HarmBenchClaude 3 Sonnet by 68.795.827.1+68.7
- AIR-Bench 2024Claude 3 Sonnet by 59.384.725.4+59.3
- simple_safety_testsClaude 3 Sonnet by 46.510053.5+46.5
- NarrativeQADBRX Instruct by 37.711.148.8+37.7
- naturalquestions_closedbookDBRX Instruct by 25.62.828.4+25.6
- GSM8KClaude 3 Sonnet by 25.292.367.1+25.2
- anthropic_red_teamClaude 3 Sonnet by 23.299.876.6+23.2
- MMLUClaude 3 Sonnet by 1478.364.3+14
- bbqClaude 3 Sonnet by 10.89079.2+10.8
- XSTestClaude 3 Sonnet by 8.485.877.4+8.4
- MATHClaude 3 Sonnet by 7.343.135.8+7.3
- OpenBookQAtie91.891tie
- AA Intelligencetie5.95.3tie
Questions people ask
Which is better, Claude 3 Sonnet or DBRX Instruct?
Claude 3 Sonnet wins all two areas where both have results: coding and reasoning. DBRX Instruct wins none.
Which is better for coding?
Claude 3 Sonnet. It wins the one coding test both models report.
How do you compare the two?
We use the 16 benchmark tests both models have published scores on. The verdict counts the 3 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.