Claude Fable 5.1 vs Claude Opus 4.8
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
Claude Fable 5.1 is the stronger all-rounder.Claude Opus 4.8 is cheaper.
Both rank among the ten best models we track in coding and agents.
Scores updated · 51 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 software70Claude Fable 5.17 of 7 tests
- ReasoningHard problems that need careful thinking60Claude Fable 5.16 of 6 tests
- AgentsCarrying out multi-step tasks on its own30Claude Fable 5.13 of 3 tests
- MathCompetition and research-level math30Claude Fable 5.13 of 3 tests
- FactsGetting facts right instead of making them up21Claude Fable 5.12 of 3 tests
- Images and chartsUnderstanding pictures, charts and video10Claude Fable 5.11 of 1 test
- Long documentsFinding answers in very long texts10Claude Fable 5.11 of 1 test
- Following instructionsDoing exactly what it is asked10Claude Fable 5.11 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 4.8 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 5.1 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+31.7points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+30.3points ahead
- Common-sense trick questionsSimpleBench+21.8points ahead
Where Claude Opus 4.8 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+33.3points ahead
Every test, side by side
All 51 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Fable 5.1
- LMArena · WebDevClaude Fable 5.1 by 185 rating points17411556+185 rating
- LiveBench · Agentic CodingClaude Fable 5.1 by 15.666.150.5+15.6
- SWE-bench ProClaude Fable 5.1 by 1281.269.2+12
- SciCodeClaude Fable 5.1 by 8.763.154.4+8.7
- Terminal-Bench 2.1Claude Fable 5.1 by 6.891.484.6+6.8
- SWE-bench MultilingualClaude Fable 5.1 by 4.789.184.4+4.7
- LiveBench · CodingClaude Fable 5.1 by 4.686.481.8+4.6
AgentsClaude Fable 5.1
- Terminal-Bench 4.0Claude Fable 5.1 by 30.35221.7+30.3
- GDPValClaude Fable 5.1 by 1562.747.7+15
- τ-Bench V3 · BankingClaude Fable 5.1 by 1347.234.2+13
ReasoningClaude Fable 5.1
- SimpleBenchClaude Fable 5.1 by 21.886.664.8+21.8
- ARC-AGI-2Claude Fable 5.1 by 17.99072.1+17.9
- Humanity's Last ExamClaude Fable 5.1 by 10.459.148.7+10.4
- CritPtClaude Fable 5.1 by 8.829.720.9+8.8
- LiveBench · ReasoningClaude Fable 5.1 by 2.591.789.2+2.5
- GPQA DiamondClaude Fable 5.1 by 1.793.792+1.7
FactsClaude Fable 5.1
- AA-Omniscience · Non-hallucinationClaude Opus 4.8 by 33.327.460.7+33.3
- AA-Omniscience · AccuracyClaude Fable 5.1 by 18.467.248.8+18.4
- SimpleQA VerifiedClaude Fable 5.1 by 17.870.853+17.8
Images and chartsClaude Fable 5.1
- LMArena · VisionClaude Fable 5.1 by 24 rating points13181294+24 rating
MathClaude Fable 5.1
- FrontierMath Tier 4Claude Fable 5.1 by 31.787.856.1+31.7
- FrontierMath Tiers 1-3 (v2)Claude Fable 5.1 by 10.290.280+10.2
- LiveBench · MathematicsClaude Fable 5.1 by 2.79794.3+2.7
Following instructionsClaude Fable 5.1
- LiveBench · Instruction FollowingClaude Fable 5.1 by 17372+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 4.0Claude Fable 5.1 by 32.255.823.6+32.2
- RiemannBenchClaude Fable 5.1 by 31.665.634+31.6
- HiL-Bench (Tools-allowed)Claude Fable 5.1 by 26.261.535.3+26.2
- GDPval-AA v2Claude Fable 5.1 by 260 rating points18531593+260 rating
- GDPval-AA v2 EloClaude Fable 5.1 by 242 rating points18351593+242 rating
- OSWorld 2.0Claude Fable 5.1 by 22.277.955.7+22.2
- SWE-bench MultimodalClaude Fable 5.1 by 16.354.738.4+16.3
- Program BenchClaude Fable 5.1 by 15.787.671.9+15.7
- GDPval-AA (Elo)Claude Opus 4.8 by 155 rating points17351890+155 rating
- AA Agentic IndexClaude Fable 5.1 by 15.45842.6+15.4
- AA-OmniscienceClaude Fable 5.1 by 14.743.528.8+14.7
- livebench_data_analysisClaude Fable 5.1 by 14.380.366+14.3
- HealthBench Professional (raw)Claude Fable 5.1 by 13.974.260.3+13.9
- AA IntelligenceClaude Fable 5.1 by 11.653.441.8+11.6
- livebench_languageClaude Fable 5.1 by 9.889.579.7+9.8
- DeepSWE 1.1Claude Fable 5.1 by 8.467.459+8.4
- HealthBench (raw)Claude Fable 5.1 by 7.966.758.8+7.9
- Artificial Analysis Coding IndexClaude Fable 5.1 by 7.381.674.3+7.3
- HLE (with tools)Claude Fable 5.1 by 7.16557.9+7.1
- HealthBench ProfessionalClaude Fable 5.1 by 6.362.155.8+6.3
- ARC-AGI-1Claude Fable 5.1 by 597.592.5+5
- PostTrainBenchClaude Fable 5.1 by 340.237.2+3
- OfficeQA ProClaude Fable 5.1 by 2.86966.2+2.8
- OfficeQAClaude Fable 5.1 by 2.680.277.6+2.6
- Toolathlon VerifiedClaude Opus 4.8 by 2.177.879.9+2.1
- GDP (Surge AI)tie85.184.8tie
Questions people ask
Which is better, Claude Fable 5.1 or Claude Opus 4.8?
Claude Fable 5.1 wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Claude Opus 4.8 wins none, but costs 50% less.
Which is better for coding?
Claude Fable 5.1. It wins 7 of the 7 coding tests both models report; Claude Opus 4.8 wins none.
Which is cheaper?
Claude Fable 5.1 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Opus 4.8 costs $5.00 and $25.00. That makes Claude Opus 4.8 about 50% cheaper for the same work.
How do you compare the two?
We use the 51 benchmark tests both models have published scores on. The verdict counts the 25 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.