Mercury 3 is the stronger all-rounder.
Scores updated · 6 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.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Mercury 3 costs 99% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Mercury 3 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+24.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+18.8points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+2.6points ahead
Where Claude Opus 4.5 pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 6 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningMercury 3
- ARC-AGI-2Mercury 3 by 24.237.661.8+24.2
- GPQA DiamondMercury 3 by 2.686.689.2+2.6
Following instructionsMercury 3
- IFBenchMercury 3 by 18.85876.8+18.8
- Multi-ChallengeMercury 3 by 1.45960.4+1.4
Other results2 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.
- ARC-AGI-1Mercury 3 by 6.58288.5+6.5
- DeepSearchQA (F1)Claude Opus 4.5 by 4.876.171.3+4.8
Questions people ask
Which is better, Claude Opus 4.5 or Mercury 3?
Mercury 3 wins all two areas where both have results: reasoning and following instructions. Claude Opus 4.5 wins none.
Which is cheaper?
Claude Opus 4.5 costs $5.00 per million input tokens and $25.00 per million output tokens; Mercury 3 costs $0.10 and $0.35. That makes Mercury 3 about 99% cheaper for the same work.
How do you compare the two?
We use the 6 benchmark tests both models have published scores on. The verdict counts the 4 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 2 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.