Claude Opus 4.8 vs Claude Opus 5.5
Wins 1 of 8 areas
Following instructions
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents
Claude Opus 5.5 is the stronger all-rounder.Claude Opus 4.8 is better at following instructions.
Both rank among the ten best models we track in coding and agents.
Scores updated · 44 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. 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 software06Claude Opus 5.56 of 6 tests
- ReasoningHard problems that need careful thinking05Claude Opus 5.55 of 5 tests
- AgentsCarrying out multi-step tasks on its own03Claude Opus 5.53 of 3 tests
- MathCompetition and research-level math03Claude Opus 5.53 of 3 tests
- FactsGetting facts right instead of making them up12Claude Opus 5.52 of 3 tests
- Images and chartsUnderstanding pictures, charts and video01Claude Opus 5.51 of 1 test
- Long documentsFinding answers in very long texts01Claude Opus 5.51 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Opus 4.81 of 1 test
Images and charts, long documents 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.
Claude Opus 5.5 costs 20% 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 Opus 5.5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+38.9points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+37.9points ahead
- Common-sense trick questionsSimpleBench+23.6points ahead
Where Claude Opus 4.8 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+19.3points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+6.3points ahead
Every test, side by side
All 44 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 5.5
- LMArena · WebDevClaude Opus 5.5 by 258 rating points15561814+258 rating
- LiveBench · Agentic CodingClaude Opus 5.5 by 21.250.571.7+21.2
- SWE-bench ProClaude Opus 5.5 by 20.769.289.9+20.7
- SciCodeClaude Opus 5.5 by 12.554.466.9+12.5
- SWE-bench MultilingualClaude Opus 5.5 by 9.584.493.9+9.5
- LiveBench · CodingClaude Opus 5.5 by 7.581.889.3+7.5
AgentsClaude Opus 5.5
- Terminal-Bench 4.0Claude Opus 5.5 by 37.921.759.6+37.9
- GDPValClaude Opus 5.5 by 20.647.768.3+20.6
- BrowseCompClaude Opus 5.5 by 6.584.390.8+6.5
ReasoningClaude Opus 5.5
- SimpleBenchClaude Opus 5.5 by 23.664.888.4+23.6
- ARC-AGI-2Claude Opus 5.5 by 19.672.191.7+19.6
- Humanity's Last ExamClaude Opus 5.5 by 12.748.761.4+12.7
- CritPtClaude Opus 5.5 by 10.820.931.7+10.8
- LiveBench · ReasoningClaude Opus 5.5 by 389.292.2+3
FactsClaude Opus 5.5
- AA-Omniscience · Non-hallucinationClaude Opus 4.8 by 19.360.741.4+19.3
- SimpleQA VerifiedClaude Opus 5.5 by 19.25372.2+19.2
- AA-Omniscience · AccuracyClaude Opus 5.5 by 17.448.866.2+17.4
Images and chartsClaude Opus 5.5
- LMArena · VisionClaude Opus 5.5 by 29 rating points12951324+29 rating
MathClaude Opus 5.5
- FrontierMath Tier 4Claude Opus 5.5 by 38.956.195+38.9
- FrontierMath Tiers 1-3 (v2)Claude Opus 5.5 by 11.28091.2+11.2
- LiveBench · MathematicsClaude Opus 5.5 by 2.894.397.1+2.8
Following instructionsClaude Opus 4.8
- LiveBench · Instruction FollowingClaude Opus 4.8 by 6.37265.7+6.3
Other results21 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 4.0Claude Opus 5.5 by 42.823.666.4+42.8
- RiemannBenchClaude Opus 5.5 by 35.63469.6+35.6
- OSWorld 2.0Claude Opus 5.5 by 26.155.781.8+26.1
- HiL-Bench (Tools-allowed)Claude Opus 5.5 by 23.735.359+23.7
- SWE-bench MultimodalClaude Opus 5.5 by 2338.461.4+23
- Program BenchClaude Opus 5.5 by 19.371.991.2+19.3
- AA-OmniscienceClaude Opus 5.5 by 17.628.846.4+17.6
- AA IntelligenceClaude Opus 5.5 by 15.841.857.6+15.8
- DeepSWE 1.1Claude Opus 5.5 by 15.25974.2+15.2
- livebench_data_analysisClaude Opus 5.5 by 14.36680.3+14.3
- PostTrainBenchClaude Opus 5.5 by 12.137.249.3+12.1
- HealthBench ProfessionalClaude Opus 5.5 by 9.855.865.6+9.8
- HLE (with tools)Claude Opus 5.5 by 9.857.967.7+9.8
- GDPval-AA (Elo)Claude Opus 4.8 by 70 rating points18901820+70 rating
- livebench_languageClaude Opus 5.5 by 6.679.786.3+6.6
- ARC-AGI-1Claude Opus 5.5 by 592.597.5+5
- Toolathlon VerifiedClaude Opus 4.8 by 2.179.977.8+2.1
- Legal Agent BenchmarkClaude Opus 4.8 by 1.7108.3+1.7
- OfficeQA ProClaude Opus 5.5 by 1.566.267.7+1.5
- HealthBenchClaude Opus 5.5 by 1.359.360.6+1.3
- OfficeQAClaude Opus 5.5 by 1.377.678.9+1.3
Questions people ask
Which is better, Claude Opus 4.8 or Claude Opus 5.5?
Claude Opus 5.5 wins seven of the eight areas we test: coding, agents, reasoning, facts, images and charts, math and long documents. Claude Opus 4.8 wins following instructions.
Which is better for coding?
Claude Opus 5.5. It wins 6 of the 6 coding tests both models report; Claude Opus 4.8 wins none.
Which is cheaper?
Claude Opus 4.8 costs $5.00 per million input tokens and $25.00 per million output tokens; Claude Opus 5.5 costs $4.00 and $20.00. That makes Claude Opus 5.5 about 20% cheaper for the same work.
How do you compare the two?
We use the 44 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 21 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.