Claude Fable 5 vs GPT-5.2
Wins 6 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math
Wins 0 of 8 areas
—
Claude Fable 5 is the stronger all-rounder.GPT-5.2 is cheaper.
Scores updated · 44 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 thinking60Claude Fable 56 of 6 tests
- AgentsCarrying out multi-step tasks on its own50Claude Fable 55 of 5 tests
- FactsGetting facts right instead of making them up30Claude Fable 53 of 3 tests
- MathCompetition and research-level math30Claude Fable 53 of 3 tests
- Images and chartsUnderstanding pictures, charts and video20Claude Fable 52 of 2 tests
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
- Following instructionsDoing exactly what it is asked11Even1 each
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.
GPT-5.2 costs 74% 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
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+50.1points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+36.3points ahead
Where GPT-5.2 pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+11.9points ahead
Every test, side by side
All 44 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 50.18029.9+50.1
- LMArena · WebDevClaude Fable 5 by 210 rating points16251415+210 rating
- SWE-bench MultilingualClaude Fable 5 by 19.986.666.7+19.9
- Terminal-Bench HardClaude Fable 5 by 15.962.947+15.9
- SWE-bench VerifiedClaude Fable 5 by 159580+15
- LiveBench · Agentic CodingClaude Fable 5 by 11.962.250.3+11.9
- LiveBench · CodingClaude Fable 5 by 9.98676.1+9.9
- SciCodeClaude Fable 5 by 8.96152.1+8.9
AgentsClaude Fable 5
- τ-Bench V3 · BankingClaude Fable 5 by 2738.111.1+27
- MCP AtlasClaude Fable 5 by 24.184.760.6+24.1
- BrowseCompClaude Fable 5 by 21.687.465.8+21.6
- AA ApexAgentsClaude Fable 5 by 20.343.323+20.3
- GDPValClaude Fable 5 by 7.355.648.3+7.3
ReasoningClaude Fable 5
- ARC-AGI-2Claude Fable 5 by 36.389.252.9+36.3
- SimpleBenchClaude Fable 5 by 36.181.945.8+36.1
- Humanity's Last ExamClaude Fable 5 by 17.855.537.7+17.8
- CritPtClaude Fable 5 by 1728.611.6+17
- LiveBench · ReasoningClaude Fable 5 by 6.589.783.2+6.5
- GPQA DiamondClaude Fable 5 by 2.392.690.3+2.3
FactsClaude Fable 5
- SimpleQA VerifiedClaude Fable 5 by 33.670.737.1+33.6
- AA-Omniscience · AccuracyClaude Fable 5 by 21.165.344.3+21.1
- AA-Omniscience · Non-hallucinationClaude Fable 5 by 17.636.418.8+17.6
Images and chartsClaude Fable 5
- MMMU-ProClaude Fable 5 by 9.684.274.6+9.6
- LMArena · VisionClaude Fable 5 by 79 rating points13251246+79 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 19.68767.4+19.6
- LiveBench · MathematicsClaude Fable 5 by 2.89693.2+2.8
Following instructionsEven
- LiveBench · Instruction FollowingClaude Fable 5 by 1475.861.8+14
- IFBenchGPT-5.2 by 11.963.575.4+11.9
Other results14 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.
- GDPval-AA (Elo)Claude Fable 5 by 470 rating points19321462+470 rating
- AA-OmniscienceClaude Fable 5 by 44.243.3-0.9+44.2
- Artificial Analysis Coding IndexClaude Fable 5 by 27.876.548.7+27.8
- PaperBenchClaude Fable 5 by 25.188.863.7+25.1
- DeepSearchQA (F1)Claude Fable 5 by 22.994.271.3+22.9
- AA IntelligenceClaude Fable 5 by 19.249.630.4+19.2
- HLE (with tools)Claude Fable 5 by 18.363.845.5+18.3
- ToolathlonClaude Fable 5 by 15.461.746.3+15.4
- τ²-Bench Telecom (AA run)Claude Fable 5 by 13.798.584.8+13.7
- ARC-AGI-1Claude Fable 5 by 12.398.586.2+12.3
- livebench_languageClaude Fable 5 by 10.990.779.8+10.9
- AA Agentic IndexGPT-5.2 by 9.25160.2+9.2
- LiveBenchClaude Fable 5 by 3.578.374.8+3.5
- livebench_data_analysisClaude Fable 5 by 2.380.578.2+2.3
Questions people ask
Which is better, Claude Fable 5 or GPT-5.2?
Claude Fable 5 wins six of the eight areas we test: coding, agents, reasoning, facts, images and charts and math. GPT-5.2 wins none, but costs 74% less. They are level on long documents and following instructions.
Which is better for coding?
Claude Fable 5. It wins 8 of the 8 coding tests both models report; GPT-5.2 wins none.
Which is cheaper?
Claude Fable 5 costs $10.00 per million input tokens and $50.00 per million output tokens; GPT-5.2 costs $1.75 and $14.00. That makes GPT-5.2 about 74% cheaper for the same work.
How do you compare the two?
We use the 44 benchmark tests both models have published scores on. The verdict counts the 30 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.