Claude Sonnet 4 vs DeepSeek-R1
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 6 areas
—
Claude Sonnet 4 is the stronger all-rounder.DeepSeek-R1 is cheaper.
Scores updated · 41 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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.
- CodingWriting and fixing software40Claude Sonnet 44 of 4 tests
- ReasoningHard problems that need careful thinking20Claude Sonnet 42 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own20Claude Sonnet 42 of 2 tests
- Following instructionsDoing exactly what it is asked20Claude Sonnet 42 of 2 tests
- FactsGetting facts right instead of making them up21Claude Sonnet 42 of 3 tests
- Long documentsFinding answers in very long texts10Claude Sonnet 41 of 1 test
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.
DeepSeek-R1 costs 67% 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 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+61.2points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+25points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+23.5points ahead
Where DeepSeek-R1 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+7.8points ahead
Every test, side by side
All 41 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4
- Terminal-Bench HardClaude Sonnet 4 by 2531.16.1+25
- SWE-bench VerifiedClaude Sonnet 4 by 23.572.749.2+23.5
- Terminal-Bench 2.1Claude Sonnet 4 by 17.236.319.1+17.2
- SciCodeClaude Sonnet 4 by 1.74038.3+1.7
AgentsClaude Sonnet 4
- τ-Bench V3 · BankingClaude Sonnet 4 by 10.316.76.4+10.3
- GDPValClaude Sonnet 4 by 9.79.70+9.7
ReasoningClaude Sonnet 4
- GPQA DiamondClaude Sonnet 4 by 6.977.770.8+6.9
- Humanity's Last ExamClaude Sonnet 4 by 2.210.78.5+2.2
- CritPttie0.30.6tie
FactsClaude Sonnet 4
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 61.270.99.7+61.2
- AA-Omniscience · AccuracyDeepSeek-R1 by 7.822.730.5+7.8
- Vectara HHEM hallucination ratelower is betterClaude Sonnet 4 by 110.311.3+1
Following instructionsClaude Sonnet 4
- Multi-ChallengeClaude Sonnet 4 by 16.457.140.7+16.4
- IFBenchClaude Sonnet 4 by 15.754.739+15.7
Other results26 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 by 53.264.611.4+53.2
- HarmBenchClaude Sonnet 4 by 50.19847.9+50.1
- AIME 2024DeepSeek-R1 by 36.443.479.8+36.4
- AIR-Bench 2024Claude Sonnet 4 by 35.488.352.9+35.4
- AA-OmniscienceClaude Sonnet 4 by 32.40.2-32.2+32.4
- ARC-AGI-1Claude Sonnet 4 by 24.24015.8+24.2
- CNMO 2024DeepSeek-R1 by 18.460.478.8+18.4
- Aider-PolyglotClaude Sonnet 4 by 17.470.753.3+17.4
- SimpleQADeepSeek-R1 by 14.215.930.1+14.2
- Artificial Analysis Coding IndexClaude Sonnet 4 by 1337.624.6+13
- AA IntelligenceClaude Sonnet 4 by 7.518.911.4+7.5
- vectara_avg_summary_lengthClaude Sonnet 4 by 52.3 rating points145.893.5+52.3 rating
- LiveCodeBenchClaude Sonnet 4 by 568.563.5+5
- ZebraLogicDeepSeek-R1 by 573.778.7+5
- IFEvalClaude Sonnet 4 by 4.387.683.3+4.3
- MATH-500 (EM)DeepSeek-R1 by 3.39497.3+3.3
- XSTestClaude Sonnet 4 by 2.296.694.4+2.2
- anthropic_red_teamClaude Sonnet 4 by 299.297.2+2
- simple_safety_testsClaude Sonnet 4 by 210098+2
- vectara_answer_rateClaude Sonnet 4 by 1.698.697+1.6
- vectara_factual_consistencyClaude Sonnet 4 by 189.788.7+1
- MMLUtie91.590.8tie
- MMLU-Reduxtie93.692.9tie
- AIME 2025tie70.570tie
- MMLU-Protie83.784tie
- bbqtie96.696.6tie
Questions people ask
Which is better, Claude Sonnet 4 or DeepSeek-R1?
Claude Sonnet 4 wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. DeepSeek-R1 wins none, but costs 67% less.
Which is better for coding?
Claude Sonnet 4. It wins 4 of the 4 coding tests both models report; DeepSeek-R1 wins none.
Which is cheaper?
Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens; DeepSeek-R1 costs $2.00 and $4.00. That makes DeepSeek-R1 about 67% cheaper for the same work.
How do you compare the two?
We use the 41 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 26 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.