Claude Sonnet 4.5 vs o3-mini
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Long documents
Wins 0 of 7 areas
—
Claude Sonnet 4.5 is the stronger all-rounder.o3-mini is cheaper.
Scores updated · 37 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- ReasoningHard problems that need careful thinking40Claude Sonnet 4.54 of 5 tests · 1 tie
- CodingWriting and fixing software40Claude Sonnet 4.54 of 4 tests
- FactsGetting facts right instead of making them up30Claude Sonnet 4.53 of 3 tests
- AgentsCarrying out multi-step tasks on its own20Claude Sonnet 4.52 of 3 tests · 1 tie
- MathCompetition and research-level math20Claude Sonnet 4.52 of 2 tests
- Long documentsFinding answers in very long texts10Claude Sonnet 4.51 of 1 test
- Following instructionsDoing exactly what it is asked11Even1 each
Long documents rests on a single test.
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.
o3-mini costs 69% 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 Claude Sonnet 4.5 pulls ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+51.3points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+32points ahead
- Common-sense trick questionsSimpleBench+31.5points ahead
Where o3-mini pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+9.8points ahead
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4.5
- Terminal-Bench 2.1Claude Sonnet 4.5 by 51.355.84.5+51.3
- Terminal-Bench HardClaude Sonnet 4.5 by 29.535.66.1+29.5
- SWE-bench VerifiedClaude Sonnet 4.5 by 27.977.249.3+27.9
- SciCodeClaude Sonnet 4.5 by 2.945.742.8+2.9
AgentsClaude Sonnet 4.5
- GDPValClaude Sonnet 4.5 by 20.520.50+20.5
- τ-Bench V3 · BankingClaude Sonnet 4.5 by 19.324.55.2+19.3
- Terminal-Bench 4.0tie00tie
ReasoningClaude Sonnet 4.5
- SimpleBenchClaude Sonnet 4.5 by 31.554.322.8+31.5
- ARC-AGI-2Claude Sonnet 4.5 by 10.613.63+10.6
- GPQA DiamondClaude Sonnet 4.5 by 6.183.477.3+6.1
- Humanity's Last ExamClaude Sonnet 4.5 by 5.817.812+5.8
- CritPttie1.10.3tie
FactsClaude Sonnet 4.5
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.5 by 3250.918.9+32
- SimpleQA VerifiedClaude Sonnet 4.5 by 15.430.715.3+15.4
- AA-Omniscience · AccuracyClaude Sonnet 4.5 by 11.632.921.3+11.6
MathClaude Sonnet 4.5
- FrontierMath Tiers 1-3 (v2)Claude Sonnet 4.5 by 5.323.918.6+5.3
- FrontierMath Tier 4Claude Sonnet 4.5 by 2.42.40+2.4
Long documentsClaude Sonnet 4.5
- AA-LCRClaude Sonnet 4.5 by 29.372.343+29.3
Following instructionsEven
- Multi-ChallengeClaude Sonnet 4.5 by 15.455.339.9+15.4
- IFBencho3-mini by 9.857.367.1+9.8
Other results17 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.
- τ²-Bench Telecom (AA run)Claude Sonnet 4.5 by 46.878.131.3+46.8
- AA-OmniscienceClaude Sonnet 4.5 by 42.5-0.1-42.6+42.5
- HMMT Feb. 2025Claude Sonnet 4.5 by 39.267.528.3+39.2
- AIME 2025Claude Sonnet 4.5 by 38.78748.3+38.7
- TAU-bench (airline)Claude Sonnet 4.5 by 37.67032.4+37.6
- Artificial Analysis Coding IndexClaude Sonnet 4.5 by 35.852.116.3+35.8
- ARC-AGI-1Claude Sonnet 4.5 by 29.263.734.5+29.2
- TAU-bench (retail)Claude Sonnet 4.5 by 28.686.257.6+28.6
- AIR-Bench 2024Claude Sonnet 4.5 by 14.989.874.9+14.9
- AA IntelligenceClaude Sonnet 4.5 by 9.720.711+9.7
- LiveCodeBencho3-mini by 6.77177.7+6.7
- bbqClaude Sonnet 4.5 by 4.898.994.1+4.8
- HarmBencho3-mini by 3.29295.2+3.2
- XSTestClaude Sonnet 4.5 by 2.396.494.1+2.3
- simple_safety_testsClaude Sonnet 4.5 by 110099+1
- anthropic_red_teamtie98.298.6tie
- frontiermath_tier_4_v1tie4.24.2tie
Questions people ask
Which is better, Claude Sonnet 4.5 or o3-mini?
Claude Sonnet 4.5 wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and long documents. o3-mini wins none, but costs 69% less. They are level on following instructions.
Which is better for coding?
Claude Sonnet 4.5. It wins 4 of the 4 coding tests both models report; o3-mini wins none.
Which is cheaper?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; o3-mini costs $1.10 and $4.40. That makes o3-mini about 69% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.