Claude Fable 5 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 Fable 5 is better at following instructions.
Both rank among the ten best models we track in reasoning, coding, agents, facts and math.
Scores updated · 50 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 up02Claude Opus 5.52 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video01Claude Opus 5.51 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Claude Opus 5.51 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Fable 51 of 1 test
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 60% 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
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+17.2points ahead
- Real work tasks from 44 professionsGDPVal+12.8points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+9.9points ahead
Where Claude Fable 5 pulls ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+10.1points ahead
Every test, side by side
All 50 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 189 rating points16251814+189 rating
- SWE-bench ProClaude Opus 5.5 by 9.98089.9+9.9
- LiveBench · Agentic CodingClaude Opus 5.5 by 9.562.271.7+9.5
- SWE-bench MultilingualClaude Opus 5.5 by 7.386.693.9+7.3
- SciCodeClaude Opus 5.5 by 5.96166.9+5.9
- LiveBench · CodingClaude Opus 5.5 by 3.38689.3+3.3
AgentsClaude Opus 5.5
- Terminal-Bench 4.0Claude Opus 5.5 by 17.242.459.6+17.2
- GDPValClaude Opus 5.5 by 12.855.568.3+12.8
- BrowseCompClaude Opus 5.5 by 3.487.490.8+3.4
ReasoningClaude Opus 5.5
- SimpleBenchClaude Opus 5.5 by 6.581.988.4+6.5
- Humanity's Last ExamClaude Opus 5.5 by 5.955.561.4+5.9
- CritPtClaude Opus 5.5 by 3.128.631.7+3.1
- ARC-AGI-2Claude Opus 5.5 by 2.589.291.7+2.5
- LiveBench · ReasoningClaude Opus 5.5 by 2.589.792.2+2.5
FactsClaude Opus 5.5
- AA-Omniscience · Non-hallucinationClaude Opus 5.5 by 536.441.4+5
- SimpleQA VerifiedClaude Opus 5.5 by 1.570.772.2+1.5
- AA-Omniscience · Accuracytie65.366.2tie
Images and chartsClaude Opus 5.5
- MMMU-ProClaude Opus 5.5 by 3.584.287.7+3.5
- LMArena · Visiontie13261324tie
MathClaude Opus 5.5
- FrontierMath Tier 4Claude Opus 5.5 by 4.890.295+4.8
- FrontierMath Tiers 1-3 (v2)Claude Opus 5.5 by 4.28791.2+4.2
- LiveBench · MathematicsClaude Opus 5.5 by 1.19697.1+1.1
Following instructionsClaude Fable 5
- LiveBench · Instruction FollowingClaude Fable 5 by 10.175.865.7+10.1
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.
- Terminal-Bench-Science 0.1Claude Opus 5.5 by 34.324.759+34.3
- GDPval-AA 2.1Claude Opus 5.5 by 251 rating points15951846+251 rating
- Terminal-Bench 4.0Claude Opus 5.5 by 21.944.566.4+21.9
- OSWorld 2.0Claude Opus 5.5 by 15.766.181.8+15.7
- OSWorld 2.0 (strict)Claude Opus 5.5 by 12.636.148.7+12.6
- GDPval-AA (Elo)Claude Fable 5 by 112 rating points19321820+112 rating
- OfficeQA ProClaude Opus 5.5 by 9.857.967.7+9.8
- OSWorld 2.0 (partial)Claude Opus 5.5 by 8.972.981.8+8.9
- AA IntelligenceClaude Opus 5.5 by 849.657.6+8
- PostTrainBenchClaude Opus 5.5 by 7.941.449.3+7.9
- SWE-bench MultimodalClaude Opus 5.5 by 7.354.161.4+7.3
- Legal Agent BenchmarkClaude Fable 5 by 513.38.3+5
- Program BenchClaude Opus 5.5 by 4.986.391.2+4.9
- charxiv_rq_with_pythonClaude Opus 5.5 by 4.489.493.8+4.4
- livebench_languageClaude Fable 5 by 4.490.786.3+4.4
- DeepSWE 1.1Claude Opus 5.5 by 4.27074.2+4.2
- HLE (with tools)Claude Opus 5.5 by 3.963.867.7+3.9
- AA-OmniscienceClaude Opus 5.5 by 3.143.346.4+3.1
- HiL-Bench (Tools-allowed)Claude Opus 5.5 by 2.756.359+2.7
- ARC-AGI-1Claude Fable 5 by 198.597.5+1
- GMMLUtie93.694.3tie
- HealthBenchtie61.260.6tie
- HealthBench Professionaltie6665.6tie
- livebench_data_analysistie80.580.3tie
- MILUtie92.993.1tie
- Toolathlon Verifiedtie77.977.8tie
Questions people ask
Which is better, Claude Fable 5 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 Fable 5 wins following instructions.
Which is better for coding?
Claude Opus 5.5. It wins 6 of the 6 coding tests both models report; Claude Fable 5 wins none.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Opus 5.5 costs $4.00 and $20.00. That makes Claude Opus 5.5 about 60% cheaper for the same work.
How do you compare the two?
We use the 50 benchmark tests both models have published scores on. The verdict counts the 24 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.