BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases9

Claude Fable 5 vs GPT-5.2

Full profiles: Claude Fable 5 (Anthropic) and GPT-5.2 (OpenAI): every benchmark, price and rank.

Updated · Data from 10 sources

As of 26 Oct 2026, Claude Fable 5 (max) leads on quality (69.3 vs 60.7 on the BenchLeader Index), but GPT-5.2 (xhigh) costs 4.2× less, so it is the better buy unless you need the extra quality. Pick Claude Fable 5 for knowledge (81 vs 57); pick GPT-5.2 for instruction following (74 vs 63).

Claude Fable 5 benchmarks max Anthropic GPT-5.2 benchmarks xhigh OpenAIOpen in the comparison tool

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingClaude Fable 574 vs 63 category score; WeirdML 91.9% vs 72.2%High
Agents & tool useClaude Fable 565 vs 55 category score; Terminal-Bench Hard 62.9% vs 47.0%High
MathsClaude Fable 572 vs 57 category score; FrontierMath Tiers 1–3 87.0% vs 67.4%High
ReasoningClaude Fable 574 vs 61 category score; ARC-AGI-2 89.2% vs 52.9%High
KnowledgeClaude Fable 581 vs 57 category score; AA-Omniscience 43.3 vs -0.9High
Long documentsEither66 vs 66 category score; AA-LCR 82.3% vs 82.7%Either
Following instructionsGPT-5.274 vs 63 category score; IFBench 75.4% vs 63.5%High
Lowest costGPT-5.2$4.81 vs $20.00 per 1M blended; agentic coding session $0.161 vs $0.800–
Fastest first answerEither101.8 s (Claude Fable 5 (max)) vs 124.7 s (GPT-5.2 (xhigh)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • Claude Fable 5 (max) leads on quality: 69.3 vs 60.7.
  • Claude Fable 5 (max) is stronger in agents & tools, coding, composite, knowledge, maths, reasoning.
  • GPT-5.2 (xhigh) is stronger in instruction following, long context.
  • GPT-5.2 (xhigh) is 4.2× cheaper ($4.81 vs $20.00 per 1M blended).

At a glance

SpecClaude Fable 5GPT-5.2
DeveloperAnthropicOpenAI
Released9 Jun 202611 Dec 2025
Context window (tokens)1,000,000400,000
Compared atClaude Fable 5 (max)GPT-5.2 (xhigh)
Open weightsNoNo
Price in / out / cached input per 1M$10.00 / $50.00 / $1.00$1.75 / $14.00 / $0.175

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

CategoryClaude Fable 5 (max)GPT-5.2 (xhigh)Leader
Reasoning73.860.7Claude Fable 5
Coding73.862.9Claude Fable 5
Agents & tools65.054.9Claude Fable 5
Maths72.456.8Claude Fable 5
Knowledge81.057.1Claude Fable 5
Instruction following63.173.6GPT-5.2
Long context66.366.4level
Composite85.063.5Claude Fable 5

Where sources disagree

Across the 23 benchmarks both have, Claude Fable 5 leads 19, GPT-5.2 leads 4. Every overall index that rates both puts Claude Fable 5 ahead.

Overall indexClaude Fable 5 (max)GPT-5.2 (xhigh)Leader
BenchLeader Index69.360.7Claude Fable 5
AA Intelligence Index v4.3.249.630.4Claude Fable 5
Arena Text1503†1437Claude Fable 5

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.

MeasureClaude Fable 5 (max)GPT-5.2 (xhigh)
Blended price per 1M tokens$20.00$4.81
Input / output per 1M$10.00 / $50.00$1.75 / $14.00
Cached input per 1M$1.00$0.175
Price per task: chat reply$0.019$0.0049
Price per task: summarize a 30-page report$0.150$0.029
Price per task: code edit$0.135$0.032
Price per task: agentic coding session$0.800$0.161
Price per task: structured extraction$0.030$0.0063
Cost to run a full benchmark suite$8.75–
Output speed67 tok/s81 tok/s
Time to first answer101.8 s124.7 s
Time to a complete answer109.3 s130.9 s

Each model is compared at its best-ranked reasoning-effort setting. Claude Fable 5 (max) scores 69.3 and takes 102 s to its first answer token; Claude Fable 5 scores 66.6 in 4 s, Claude Fable 5 (high) scores 64.4 in 4 s. GPT-5.2 (xhigh) scores 60.7 and takes 125 s to its first answer token; GPT-5.2 scores 58.0 in 2 s, GPT-5.2 (medium) scores 57.6 in 2 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

147 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublisherClaude Fable 5 (max)GPT-5.2 (xhigh)
BenchLeader IndexBenchLeader69.360.7
Agents & tools scoreBenchLeader65.054.9
Coding scoreBenchLeader73.862.9
Composite scoreBenchLeader85.063.5
Instruction following scoreBenchLeader63.173.6
Knowledge scoreBenchLeader81.057.1
Long context scoreBenchLeader66.366.4
Maths scoreBenchLeader72.456.8
Reasoning scoreBenchLeader73.860.7
Blended price $/MBenchLeader$20.00$4.81
Output speedBenchLeader67 tok/s81 tok/s
Time to first answerBenchLeader101.8 s124.7 s
Context windowBenchLeader1,000,000 tokens400,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub85.9%91.4%
Humanity's Last ExamScale AI / CAIS–27.8%†
SimpleBenchSimpleBench81.9%45.8%†
Arena Hard PromptsArena (LMArena)1529†1497†
LiveBench 1.0 ReasoningLiveBench89.7%83.2%†
GPQA Diamond (AA)Artificial Analysis92.6%90.3%
Humanity's Last Exam (AA)Artificial Analysis55.5%37.7%
GPQA Diamond (Vals)Vals AI93.2%†91.7%
EnigmaEvalScale AI SEAL39.3%†–
Kagi LLM BenchmarkKagi LLM Benchmark88.8%†73.3%†
ARC-AGI-1ARC Prize98.5%86.2%
ARC-AGI-2ARC Prize89.2%52.9%
CritPtArtificial Analysis28.6%11.6%
SciPredictScale AI SEAL–20.6%†
Chess PuzzlesEpoch AI Benchmarking Hub47.0%48.0%
Mystery Game PuzzlesEpoch AI Benchmarking Hub52.0%22.0%†
CL-benchEpoch AI Benchmarking Hub–18.2%†
LMCAEpoch AI Benchmarking Hub60.3%43.9%
DTBenchEpoch AI Benchmarking Hub98.4%90.9%
ForecastBenchForecastBench66.0%64.2%†
BTF-3Epoch AI Benchmarking Hub13.0%†–
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub–73.8%†
SciCodeSciCode61.0%–
WeirdMLWeirdML91.9%72.2%
FrontierCodeCognition53.5%†–
Show all 147 benchmarks
MetricPublisherClaude Fable 5 (max)GPT-5.2 (xhigh)
Coding
GSO-BenchGSO-Bench78.4%†27.4%†
Arena CodingArena (LMArena)1551†1515†
Arena WebDevArena (LMArena)1624†1417†
LiveBench 1.0 CodingLiveBench86.0%76.1%†
LiveCodeBenchVals AI89.8%†85.4%
SWE-bench (Vals)Vals AI95.0%†75.8%
SWE-Bench ProScale AI SEAL–29.9%†
SWE-bench Verified (bash only)SWE-bench–69.0%†
SWE-bench Verified (any scaffold)SWE-bench–72.8%†
Code MigrationVals AI55.1%†–
ProgramBenchVals AI2.0%†–
Vibe Code Bench v1.1Vals AI90.3%†53.5%
SWE-Bench Pro (private)Scale AI SEAL–29.9%†
SWE Atlas: Codebase QnAScale AI SEAL39.0%†–
SWE Atlas: RefactoringScale AI SEAL54.8%†–
SWE Atlas: Test WritingScale AI SEAL55.6%†–
MirrorCodeEpoch AI Benchmarking Hub63.9%†–
Surface Evolver BenchEpoch AI Benchmarking Hub95.0%†–
DeepSWE v1.1Epoch AI Benchmarking Hub69.7%–
ALE-BenchEpoch AI Benchmarking Hub2041.3†1249.8†
AlgoTuneAlgoTune–2.0†
FrontierSWEEpoch AI Benchmarking Hub47.0%–
Terminal-Bench 4.0 (AA)Artificial Analysis42.4%–
Terminal-Bench 2.1 (AA)Artificial Analysis84.6%–
Startrise FrontendStartrise73.8%†–
Agents & tools
Terminal-BenchTerminal-Bench44.5%64.9%†
GDPvalOpenAI–49.7%†
Remote Labor IndexScale AI / CAIS15.8%†2.1%†
APEX-AgentsMercor63.6%†–
Arena AgentArena (LMArena)7.8†–
LiveBench Agentic CodingLiveBench62.2%50.3%†
Terminal-Bench HardArtificial Analysis62.9%47.0%
τ²-Bench Telecom (AA)Artificial Analysis98.5%84.8%
Terminal-Bench 2.1 (Vals)Vals AI80.5%†–
MCP AtlasScale AI SEAL83.3%†67.6%
HiL-BenchScale AI SEAL56.3%†–
BFCL OverallBerkeley Function Calling Leaderboard–55.9%†
τ²-benchτ²-bench–84.8%†
τ³-bench Bankingτ²-bench39.7%32.2%†
τ³-Banking (AA)Artificial Analysis38.1%11.1%†
Analyst Agent (AA)Artificial Analysis48.8%–
Finance Agent v2Vals AI56.3%†–
Harvey's Legal Agent Benchmark v1Vals AI11.3%†–
Legal Research BenchVals AI49.5%†–
Poker AgentVals AI–1131.8%†
Tax Agent BenchVals AI38.3%†–
Terminal-Bench 2.0 (Vals)Vals AI–51.7%†
Terminal-Bench 4.0 (Vals)Vals AI41.4%†–
Terminal-Bench ScienceVals AI15.7%†–
Web Search IndexVals AI46.9%†–
Arena SearchArena (LMArena)1229†–
DeepResearch BenchDeepResearch Bench–41.1%†
PostTrainBenchEpoch AI Benchmarking Hub41.8%–
METR Time HorizonsMETR–75.3%†
GBAEvalEpoch AI Benchmarking Hub74.5%†–
Vending-Bench 2Andon Labs4966.63591.3†
AutomationBenchArtificial Analysis54.1%–
GDP.pdf (AA)Artificial Analysis24.0%–
EnterpriseOps-GymArtificial Analysis51.1%–
AA-Briefcase v1.1Artificial Analysis1537–
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub87.0%67.4%
FrontierMath Tier 4Epoch AI Benchmarking Hub90.2%31.7%
OTIS Mock AIMEEpoch AI Benchmarking Hub99.7%96.1%
ProofBenchVals AI95.0%15.0%
LiveBench MathematicsLiveBench96.0%93.2%†
AIME (Vals)Vals AI–96.9%
MGSMVals AI–94.0%
AIME 2026MathArena–98.3%†
HMMT February 2026MathArena–97.0%†
MathArena ApexMathArena–13.5%†
Arena MathsArena (LMArena)1521†1454†
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub70.7%†37.1%
LiveBench 1.0 Data AnalysisLiveBench80.5%78.2%†
AA-OmniscienceArtificial Analysis43.3-0.9
MMLU-ProVals AI91.5%†86.2%
LegalBenchVals AI88.6%†82.8%
CorpFinVals AI71.8%†65.9%
TaxEvalVals AI76.9%†75.8%
MedQAVals AI–94.1%
PRBench FinanceScale AI SEAL53.9%†–
PRBench LegalScale AI SEAL52.6%†–
MultiNRCScale AI SEAL–42.2%†
CaseLaw v2Vals AI–66.0%
Excel Modeling BenchmarkVals AI73.7%†–
MedCodeVals AI56.1%†49.8%
MedScribeVals AI88.5%†84.4%
Public Benefits Bench v1.1Vals AI70.4%†–
DrugDiscoveryBenchScale AI SEAL–29.3%
EBR-benchEpoch AI Benchmarking Hub39.5%23.0%
GDP.pdfEpoch AI Benchmarking Hub29.8%–
AA-Omniscience: accuracyArtificial Analysis65.3%44.3%
AA-Omniscience: non-hallucinationArtificial Analysis36.4%18.8%
Global-MMLU-Lite (AA)Artificial Analysis–89.8%†
Instruction following
LiveBench LanguageLiveBench90.7%79.8%†
LiveBench Instruction FollowingLiveBench75.8%61.8%†
IFBenchArtificial Analysis63.5%75.4%
TutorBenchScale AI SEAL–53.5%†
Arena Instruction FollowingArena (LMArena)1508†1457†
Human preference
Arena TextArena (LMArena)1503†1476†
EQ-Bench 4EQ-Bench1310†–
Arena Creative WritingArena (LMArena)1501†1433†
Arena Multi-turnArena (LMArena)1517†1495†
Arena Longer QueriesArena (LMArena)1522†1472†
Multimodal
Arena VisionArena (LMArena)1325†1268†
MMMU-ProArtificial Analysis–74.6%†
VISTAScale AI SEAL–46.6%†
MMMU-Pro (Vals)Vals AI89.3%†86.7%
MortgageTaxVals AI68.9%†67.1%
SAGEVals AI51.9%†49.3%
VPCTEpoch AI Benchmarking Hub–84.0%
Blueprint-Bench 2Epoch AI Benchmarking Hub38.6%†–
Long context
AA-LCRArtificial Analysis82.3%82.7%
Arena DocumentArena (LMArena)1499†1425†
MLCRArtificial Analysis20.6%–
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub163.4†153.7†
LiveBench 1.0LiveBench83.0%74.6%†
AA Intelligence Index v4.3.2Artificial Analysis49.630.4
Vals RSI IndexVals AI24.3%15.8%
Vals Multimodal IndexVals AI74.2%†–
Safety & honesty
FORTRESSScale AI SEAL–17.5%†
MASKScale AI SEAL–86.7%†
PropensityBenchScale AI SEAL–34.4%†

Data as of 2026-10-26. Best configuration of each model; every score links to its source on the model pages.

What changed

  • Claude Fable 5 (max): 1 new result (Vals RSI Index).
  • GPT-5.2 (xhigh): 1 new result (Vals RSI Index).
  • Claude Fable 5 (max): 1 score revision.
  • Claude Fable 5 (max): 1 new result (ForecastBench).
  • Claude Fable 5 (max): 4 new results (GDPval-AA v2.1, AA-Omniscience: accuracy, AA-Omniscience: non-hallucination…).
  • Claude Fable 5 (max): 1 score revision.
  • GPT-5.2 (xhigh): 1 score revision.
  • GPT-5.2 (xhigh): 1 new result (DrugDiscoveryBench).
  • Claude Fable 5 (max): BenchLeader Index 64.6 → 69.1.
  • Claude Fable 5 (max): 23 new results (AA Intelligence Index v4.3.2, GPQA Diamond (AA), Humanity's Last Exam (AA)…).

Rankings: Best LLM overall · Best LLM for coding · Best LLM for agents · Best cheap LLMs

Claude Fable 5 vs GPT-5.2: questions

Is Claude Fable 5 better than GPT-5.2?
Claude Fable 5 (max) leads on quality: 69.3 vs 60.7. The BenchLeader Index combines every independent quality benchmark; Claude Fable 5 (max) is ahead overall as of 2026-10-26, but check the category scores for your use.
Which is cheaper, Claude Fable 5 or GPT-5.2, and will that change?
GPT-5.2 is cheaper today: $4.81 against $20.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is Claude Fable 5 better than GPT-5.2 for coding?
Claude Fable 5 scores higher in coding (74 vs 63 on the category index, where 50 is average). On WeirdML, the top shared coding benchmark, it is 91.9% for Claude Fable 5 against 72.2% for GPT-5.2.
Is Claude Fable 5 better than GPT-5.2 for agents and tool use?
Claude Fable 5 scores higher in agentic tasks (65 vs 55 on the category index, where 50 is average). On Terminal-Bench Hard, the top shared agents & tools benchmark, it is 62.9% for Claude Fable 5 against 47.0% for GPT-5.2.
Which is better for long documents?
They are level on long-context benchmarks (66 vs 66). Claude Fable 5 accepts more context: 1,000,000 against 400,000 tokens for GPT-5.2.
Which has lower latency?
Claude Fable 5 (max) answers first: 101.8 s against 124.7 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Which is faster, Claude Fable 5 or GPT-5.2?
GPT-5.2 streams faster: 81 against 67 output tokens per second.
Do independent benchmarks agree on Claude Fable 5 vs GPT-5.2?
Across the 23 benchmarks both have, Claude Fable 5 leads 19, GPT-5.2 leads 4. Every overall index that rates both puts Claude Fable 5 ahead.
When was Claude Fable 5 released?
Claude Fable 5 was released on 9 Jun 2026.