Claude Fable 5.1 vs Claude Fable 5.2
Wins 1 of 6 areas
Long documents
Wins 5 of 6 areas
Coding · Agents · Reasoning · Facts · Math
Claude Fable 5.2 is the stronger all-rounder.Claude Fable 5.1 is better at long documents.
Both rank among the ten best models we track in math.
Scores updated · 35 tests both models report · How we compare
Where each one wins
Tests won in each of the six areas where both have results. 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 software03Claude Fable 5.23 of 3 tests
- ReasoningHard problems that need careful thinking03Claude Fable 5.23 of 3 tests
- MathCompetition and research-level math02Claude Fable 5.22 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Claude Fable 5.21 of 1 test
- FactsGetting facts right instead of making them up01Claude Fable 5.21 of 1 test
- Long documentsFinding answers in very long texts10Claude Fable 5.11 of 1 test
Agents, facts and long documents 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 Fable 5.2 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 Fable 5.2 pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+23.2points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+15.2points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+12.2points ahead
Where Claude Fable 5.1 pulls ahead
- Reasons across sets of long documentsAA-LCR+2points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Fable 5.2
- SWE-bench ProClaude Fable 5.2 by 11.881.293+11.8
- SWE-bench MultilingualClaude Fable 5.2 by 7.589.196.6+7.5
- SciCodeClaude Fable 5.2 by 4.663.167.7+4.6
ReasoningClaude Fable 5.2
- Humanity's Last ExamClaude Fable 5.2 by 15.259.174.3+15.2
- CritPtClaude Fable 5.2 by 5.729.735.4+5.7
- ARC-AGI-2Claude Fable 5.2 by 2.59092.5+2.5
MathClaude Fable 5.2
- FrontierMath Tier 4Claude Fable 5.2 by 12.287.8100+12.2
- FrontierMath Tiers 1-3 (v2)Claude Fable 5.2 by 4.990.295.1+4.9
Other results24 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 Fable 5.2 by 3043.373.3+30
- CritPt v2 HardClaude Fable 5.2 by 2452.576.5+24
- HLE-DiamondClaude Fable 5.2 by 20.160.180.2+20.1
- SWE-bench MultimodalClaude Fable 5.2 by 19.454.774.1+19.4
- Terminal-Bench 4.0Claude Fable 5.2 by 11.855.867.6+11.8
- HealthBench ProfessionalClaude Fable 5.2 by 9.957.867.7+9.9
- HealthBench Professional length-adjustedClaude Fable 5.2 by 9.957.867.7+9.9
- GDPval-AA 2.1Claude Fable 5.2 by 96 rating points17351831+96 rating
- HLE (with tools)Claude Fable 5.2 by 9.26574.2+9.2
- OSWorld 2.1 partialClaude Fable 5.2 by 8.680.789.3+8.6
- OSWorld 2.1 offline set (strict pass rate)Claude Fable 5.2 by 747.654.6+7
- DeepSWE 1.1Claude Fable 5.2 by 6.967.474.3+6.9
- GDPval-AA v2.1 (max effort)Claude Fable 5.2 by 68 rating points17631831+68 rating
- OSWorld 2.1 offline set (partial score)Claude Fable 5.2 by 5.883.589.3+5.8
- ProgramBench (with context tokens remaining shown)Claude Fable 5.2 by 4.886.891.6+4.8
- OfficeQA ProClaude Fable 5.2 by 3.96972.9+3.9
- OfficeQAClaude Fable 5.2 by 3.280.283.4+3.2
- HealthBenchClaude Fable 5.2 by 2.457.259.6+2.4
- HealthBench length-adjustedClaude Fable 5.2 by 2.457.259.6+2.4
- GDP (Surge AI)Claude Fable 5.1 by 226.224.2+2
- GMMLUClaude Fable 5.2 by 29496+2
- Program BenchClaude Fable 5.1 by 287.685.6+2
- MILUClaude Fable 5.2 by 1.19394.1+1.1
- ARC-AGI-1tie97.597.5tie
Questions people ask
Which is better, Claude Fable 5.1 or Claude Fable 5.2?
Claude Fable 5.2 wins five of the six areas where both have results: coding, agents, reasoning, facts and math. Claude Fable 5.1 wins long documents.
Which is better for coding?
Claude Fable 5.2. It wins 3 of the 3 coding tests both models report; Claude Fable 5.1 wins none.
Which is cheaper?
Claude Fable 5.1 costs $10.00 per million input tokens and $50.00 per million output tokens; Claude Fable 5.2 costs $8.00 and $40.00. That makes Claude Fable 5.2 about 20% cheaper for the same work.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 24 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.