Claude Fable 5 vs Claude Opus 4.7
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
Claude Fable 5 is the stronger all-rounder.Claude Opus 4.7 is cheaper.
Both rank among the ten best models we track in coding.
Scores updated · 53 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 software90Claude Fable 59 of 9 tests
- AgentsCarrying out multi-step tasks on its own60Claude Fable 56 of 6 tests
- ReasoningHard problems that need careful thinking60Claude Fable 56 of 6 tests
- MathCompetition and research-level math30Claude Fable 53 of 3 tests
- Images and chartsUnderstanding pictures, charts and video20Claude Fable 52 of 2 tests
- Following instructionsDoing exactly what it is asked20Claude Fable 52 of 2 tests
- FactsGetting facts right instead of making them up21Claude Fable 52 of 3 tests
- Long documentsFinding answers in very long texts10Claude Fable 51 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.
Claude Opus 4.7 costs 50% 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+58.5points ahead
- Common-sense trick questionsSimpleBench+20.2points ahead
- Short factual questions, answered correctlySimpleQA Verified+19points ahead
Where Claude Opus 4.7 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+21.3points ahead
Every test, side by side
All 53 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 15.78064.3+15.7
- LiveBench · Agentic CodingClaude Fable 5 by 11.562.250.7+11.5
- Terminal-Bench HardClaude Fable 5 by 11.462.951.5+11.4
- SWE-bench VerifiedClaude Fable 5 by 7.49587.6+7.4
- LMArena · WebDevClaude Fable 5 by 66 rating points16241558+66 rating
- SciCodeClaude Fable 5 by 6.56154.5+6.5
- SWE-bench MultilingualClaude Fable 5 by 6.186.680.5+6.1
- LiveBench · CodingClaude Fable 5 by 3.98682.1+3.9
- Terminal-Bench 2.1Claude Fable 5 by 1.584.683.1+1.5
AgentsClaude Fable 5
- GDPValClaude Fable 5 by 12.855.642.8+12.8
- AA ApexAgentsClaude Fable 5 by 9.443.333.9+9.4
- BrowseCompClaude Fable 5 by 8.187.479.3+8.1
- MCP AtlasClaude Fable 5 by 7.484.777.3+7.4
- OSWorld-VerifiedClaude Fable 5 by 78578+7
- τ-Bench V3 · BankingClaude Fable 5 by 3.538.134.6+3.5
ReasoningClaude Fable 5
- SimpleBenchClaude Fable 5 by 20.281.961.7+20.2
- CritPtClaude Fable 5 by 16.628.612+16.6
- ARC-AGI-2Claude Fable 5 by 13.489.275.8+13.4
- Humanity's Last ExamClaude Fable 5 by 13.255.542.3+13.2
- LiveBench · ReasoningClaude Fable 5 by 2.589.787.2+2.5
- GPQA DiamondClaude Fable 5 by 1.292.691.4+1.2
FactsClaude Fable 5
- AA-Omniscience · Non-hallucinationClaude Opus 4.7 by 21.336.457.7+21.3
- SimpleQA VerifiedClaude Fable 5 by 1970.751.7+19
- AA-Omniscience · AccuracyClaude Fable 5 by 16.565.348.9+16.5
Images and chartsClaude Fable 5
- MMMU-ProClaude Fable 5 by 5.484.278.8+5.4
- LMArena · VisionClaude Fable 5 by 12 rating points13251313+12 rating
MathClaude Fable 5
- FrontierMath Tier 4Claude Fable 5 by 58.590.231.7+58.5
- FrontierMath Tiers 1-3 (v2)Claude Fable 5 by 16.88770.2+16.8
- LiveBench · MathematicsClaude Fable 5 by 3.19692.8+3.1
Following instructionsClaude Fable 5
- LiveBench · Instruction FollowingClaude Fable 5 by 9.175.866.7+9.1
- IFBenchClaude Fable 5 by 4.963.558.6+4.9
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.
- OfficeQA ProClaude Opus 4.7 by 22.757.980.6+22.7
- SWE-bench MultimodalClaude Fable 5 by 19.654.134.5+19.6
- GDPval-AA (Elo)Claude Fable 5 by 179 rating points19321753+179 rating
- AA-OmniscienceClaude Fable 5 by 1643.327.3+16
- DeepSWEClaude Fable 5 by 167054+16
- HiL-Bench (Tools-allowed)Claude Fable 5 by 14.656.341.7+14.6
- Blueprint-Bench 2Claude Fable 5 by 14.138.624.5+14.1
- livebench_languageClaude Fable 5 by 12.890.777.9+12.8
- AA Agentic IndexClaude Fable 5 by 11.55139.5+11.5
- τ²-Bench Telecom (AA run)Claude Fable 5 by 9.998.588.6+9.9
- HLE (with tools)Claude Fable 5 by 9.163.854.7+9.1
- AA IntelligenceClaude Fable 5 by 8.949.640.7+8.9
- ARC-AGI-1Claude Fable 5 by 6.598.592+6.5
- Legal Agent BenchmarkClaude Fable 5 by 6.213.37.1+6.2
- Toolathlon Avg turnsClaude Opus 4.7 by 6.119.825.9+6.1
- Finance Agent v2Claude Fable 5 by 4.856.351.5+4.8
- Artificial Analysis Coding IndexClaude Fable 5 by 2.976.573.6+2.9
- Toolathlon Pass@∞Claude Fable 5 by 2.855.652.8+2.8
- ToolathlonClaude Fable 5 by 2.461.759.3+2.4
- livebench_data_analysisClaude Fable 5 by 2.280.578.3+2.2
- LiveBenchClaude Fable 5 by 1.478.376.9+1.4
Questions people ask
Which is better, Claude Fable 5 or Claude Opus 4.7?
Claude Fable 5 wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Claude Opus 4.7 wins none, but costs 50% less.
Which is better for coding?
Claude Fable 5. It wins 9 of the 9 coding tests both models report; Claude Opus 4.7 wins none.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Opus 4.7 costs $5.00 and $25.00. That makes Claude Opus 4.7 about 50% cheaper for the same work.
How do you compare the two?
We use the 53 benchmark tests both models have published scores on. The verdict counts the 32 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.