Claude Sonnet 5 vs Sonnet 5.5
Wins 1 of 6 areas
Following instructions
Wins 5 of 6 areas
Coding · Reasoning · Facts · Images and charts · Math
Sonnet 5.5 is the stronger all-rounder.Claude Sonnet 5 is better at following instructions.
Scores updated · 33 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 software14Sonnet 5.54 of 5 tests
- ReasoningHard problems that need careful thinking03Sonnet 5.53 of 3 tests
- MathCompetition and research-level math03Sonnet 5.53 of 3 tests
- FactsGetting facts right instead of making them up01Sonnet 5.51 of 1 test
- Images and chartsUnderstanding pictures, charts and video01Sonnet 5.51 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Sonnet 51 of 1 test
Facts, images and charts and following instructions rest on a single test each.
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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Sonnet 5.5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+51.2points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+23.2points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+23.2points ahead
Where Claude Sonnet 5 pulls ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+7.1points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+3.1points ahead
Every test, side by side
All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingSonnet 5.5
- LMArena · WebDevSonnet 5.5 by 247 rating points15391786+247 rating
- SWE-bench ProSonnet 5.5 by 18.163.281.3+18.1
- SWE-bench MultilingualSonnet 5.5 by 1278.390.3+12
- LiveBench · CodingSonnet 5.5 by 10.780.791.4+10.7
- LiveBench · Agentic CodingClaude Sonnet 5 by 3.159.456.3+3.1
ReasoningSonnet 5.5
- Humanity's Last ExamSonnet 5.5 by 23.241.364.5+23.2
- SimpleBenchSonnet 5.5 by 15.360.675.9+15.3
- LiveBench · ReasoningSonnet 5.5 by 2.988.791.6+2.9
Images and chartsSonnet 5.5
- LMArena · VisionSonnet 5.5 by 15 rating points12751290+15 rating
MathSonnet 5.5
- FrontierMath Tier 4Sonnet 5.5 by 51.229.380.5+51.2
- FrontierMath Tiers 1-3 (v2)Sonnet 5.5 by 23.265.688.8+23.2
- LiveBench · MathematicsSonnet 5.5 by 3.292.996.1+3.2
Following instructionsClaude Sonnet 5
- LiveBench · Instruction FollowingClaude Sonnet 5 by 7.163.956.8+7.1
Other results19 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.
- Terminal-Bench 4.0Sonnet 5.5 by 58.212.470.6+58.2
- GDPval-AA 2.1Sonnet 5.5 by 391 rating points14491840+391 rating
- OSWorld 2.1Sonnet 5.5 by 26.95783.9+26.9
- SWE-bench MultimodalSonnet 5.5 by 26.228.154.3+26.2
- OSWorld 2.1 partialSonnet 5.5 by 23.15780.1+23.1
- OSWorld 2.1 partial scoreSonnet 5.5 by 23.15780.1+23.1
- OSWorld 2.1 strict pass rateSonnet 5.5 by 17.925.643.5+17.9
- AA IntelligenceSonnet 5.5 by 17.838.256+17.8
- DeepSWE 1.1Sonnet 5.5 by 175471+17
- livebench_data_analysisClaude Sonnet 5 by 12.271.759.5+12.2
- HealthBench ProfessionalSonnet 5.5 by 11.457.869.2+11.4
- HLE (with tools)Sonnet 5.5 by 7.157.464.5+7.1
- OfficeQA ProSonnet 5.5 by 6.259.465.6+6.2
- Legal Agent BenchmarkSonnet 5.5 by 5.95.811.7+5.9
- Toolathlon VerifiedSonnet 5.5 by 3.174.777.8+3.1
- livebench_languageSonnet 5.5 by 37578+3
- Program BenchSonnet 5.5 by 2.477.379.7+2.4
- MILUSonnet 5.5 by 2.389.391.6+2.3
- OfficeQASonnet 5.5 by 1.875.176.9+1.8
Questions people ask
Which is better, Claude Sonnet 5 or Sonnet 5.5?
Sonnet 5.5 wins five of the six areas where both have results: coding, reasoning, facts, images and charts and math. Claude Sonnet 5 wins following instructions.
Which is better for coding?
Sonnet 5.5. It wins 4 of the 5 coding tests both models report; Claude Sonnet 5 wins 1.
Which is cheaper?
Claude Sonnet 5 costs $2.00 per million input tokens and $10.00 per million output tokens; Sonnet 5.5 costs $2.00 and $10.00. For a million tokens read plus a million written, they cost about the same.
How do you compare the two?
We use the 33 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 19 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.