Claude Fable 5 vs Claude Sonnet 5
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Following instructions
Wins 0 of 8 areas
—
Claude Fable 5 is the stronger all-rounder.Claude Sonnet 5 is cheaper.
Scores updated · 54 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 software80Claude Fable 58 of 8 tests
- ReasoningHard problems that need careful thinking50Claude Fable 55 of 5 tests
- AgentsCarrying out multi-step tasks on its own40Claude Fable 54 of 5 tests · 1 tie
- MathCompetition and research-level math30Claude Fable 53 of 3 tests
- Images and chartsUnderstanding pictures, charts and video20Claude Fable 52 of 2 tests
- FactsGetting facts right instead of making them up21Claude Fable 52 of 3 tests
- Following instructionsDoing exactly what it is asked10Claude Fable 51 of 1 test
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
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 Sonnet 5 costs 80% 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 Fable 5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+60.9points ahead
- Short factual questions, answered correctlySimpleQA Verified+37points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+28.3points ahead
Where Claude Sonnet 5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+24.2points ahead
Every test, side by side
All 54 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Fable 5
- SWE-bench ProClaude Fable 5 by 16.88063.2+16.8
- SWE-bench VerifiedClaude Fable 5 by 9.89585.2+9.8
- LMArena · WebDevClaude Fable 5 by 86 rating points16251539+86 rating
- SWE-bench MultilingualClaude Fable 5 by 8.386.678.3+8.3
- SciCodeClaude Fable 5 by 6.76154.3+6.7
- LiveBench · CodingClaude Fable 5 by 5.38680.7+5.3
- Terminal-Bench 2.1Claude Fable 5 by 4.184.680.5+4.1
- LiveBench · Agentic CodingClaude Fable 5 by 2.862.259.4+2.8
AgentsClaude Fable 5
- Terminal-Bench 4.0Claude Fable 5 by 28.342.414.1+28.3
- GDPValClaude Fable 5 by 7.355.648.3+7.3
- OSWorld-VerifiedClaude Fable 5 by 3.88581.2+3.8
- BrowseCompClaude Fable 5 by 2.787.484.7+2.7
- τ-Bench V3 · Bankingtie38.137.3tie
ReasoningClaude Fable 5
- SimpleBenchClaude Fable 5 by 21.381.960.6+21.3
- Humanity's Last ExamClaude Fable 5 by 14.255.541.3+14.2
- CritPtClaude Fable 5 by 11.728.616.9+11.7
- GPQA DiamondClaude Fable 5 by 1.592.691.1+1.5
- LiveBench · ReasoningClaude Fable 5 by 189.788.7+1
FactsClaude Fable 5
- SimpleQA VerifiedClaude Fable 5 by 3770.733.7+37
- AA-Omniscience · AccuracyClaude Fable 5 by 25.365.340+25.3
- AA-Omniscience · Non-hallucinationClaude Sonnet 5 by 24.236.460.6+24.2
Images and chartsClaude Fable 5
- MMMU-ProClaude Fable 5 by 6.984.277.3+6.9
- LMArena · VisionClaude Fable 5 by 50 rating points13251275+50 rating
MathClaude Fable 5
- FrontierMath Tier 4Claude Fable 5 by 60.990.229.3+60.9
- FrontierMath Tiers 1-3 (v2)Claude Fable 5 by 21.48765.6+21.4
- LiveBench · MathematicsClaude Fable 5 by 3.19692.9+3.1
Following instructionsClaude Fable 5
- LiveBench · Instruction FollowingClaude Fable 5 by 11.975.863.9+11.9
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.
- GDP (Surge AI)Claude Sonnet 5 by 51.829.881.6+51.8
- Terminal-Bench 4.0Claude Fable 5 by 32.144.512.4+32.1
- OSWorld 2.0 (partial)Claude Fable 5 by 30.372.942.6+30.3
- AA-OmniscienceClaude Fable 5 by 26.843.316.4+26.8
- SWE-bench MultimodalClaude Fable 5 by 2654.128.1+26
- DeepSWE 1.1Claude Fable 5 by 167054+16
- livebench_languageClaude Fable 5 by 15.790.775+15.7
- Toolathlon Pass@∞Claude Fable 5 by 14.955.640.7+14.9
- GDPval-AA 2.1Claude Fable 5 by 146 rating points15951449+146 rating
- GDPval-AA v2 EloClaude Fable 5 by 139 rating points17231584+139 rating
- AA IntelligenceClaude Fable 5 by 11.449.638.2+11.4
- GDPval-AA v2Claude Fable 5 by 105 rating points17231618+105 rating
- Program BenchClaude Fable 5 by 986.377.3+9
- livebench_data_analysisClaude Fable 5 by 8.880.571.7+8.8
- HealthBench ProfessionalClaude Fable 5 by 8.26657.8+8.2
- Legal Agent BenchmarkClaude Fable 5 by 7.513.35.8+7.5
- ToolathlonClaude Fable 5 by 7.461.754.3+7.4
- AA Agentic IndexClaude Fable 5 by 6.75144.3+6.7
- HealthBench Professional (raw)Claude Fable 5 by 6.568.962.4+6.5
- HLE (with tools)Claude Fable 5 by 6.463.857.4+6.4
- Artificial Analysis Coding IndexClaude Fable 5 by 576.571.5+5
- GMMLUClaude Fable 5 by 4.493.689.2+4.4
- MILUClaude Fable 5 by 3.692.989.3+3.6
- Toolathlon VerifiedClaude Fable 5 by 3.277.974.7+3.2
- HealthBench (raw)Claude Fable 5 by 261.259.2+2
- OfficeQA ProClaude Sonnet 5 by 1.557.959.4+1.5
Questions people ask
Which is better, Claude Fable 5 or Claude Sonnet 5?
Claude Fable 5 wins seven of the eight areas we test: coding, agents, reasoning, facts, images and charts, math and following instructions. Claude Sonnet 5 wins none, but costs 80% less. They are level on long documents.
Which is better for coding?
Claude Fable 5. It wins 8 of the 8 coding tests both models report; Claude Sonnet 5 wins none.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Sonnet 5 costs $2.00 and $10.00. That makes Claude Sonnet 5 about 80% cheaper for the same work.
How do you compare the two?
We use the 54 benchmark tests both models have published scores on. The verdict counts the 28 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.