Noometry

Model comparison

Claude Fable 5 vs Claude Sonnet 4

Claude Fable 5 is the stronger model overall, scoring 66.8 to 40.8 on the Noometry Index. Claude Sonnet 4 costs 3.3× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Last verified . 34 shared benchmarks.

Claude Fable 5 Anthropic

66.8

Rank #5 Confirmed

Claude Sonnet 4 Anthropic

40.8

Rank #145 Confirmed

Summary

  • They share 34 benchmarks with published results for both. Claude Fable 5 scores higher in 10 categories and Claude Sonnet 4 in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Claude Fable 5 leads 76.8 to 22.9.
  • The biggest single-benchmark swing is ARC-AGI-2: 89.2% for Claude Fable 5 and 5.9% for Claude Sonnet 4.
  • Claude Sonnet 4 is cheaper at $3 / $15 per million input/output tokens, against $10 / $50 for Claude Fable 5.
  • Claude Fable 5 accepts more context: 1M tokens versus 200K.

Side by side

Claude Fable 5Claude Sonnet 4
ProviderAnthropicAnthropic
Noometry Index66.840.8
Released2026-06-072025-05-22
WeightsProprietaryProprietary
Context window1M200K
Max output128K64K
Input $ / M tokens$10$3
Output $ / M tokens$50$15
Results tracked6258

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Claude Fable 5 leads

Claude Fable 5: 70.6 (#4), Claude Sonnet 4: 43.5 (#88)

BenchmarkClaude Fable 5Claude Sonnet 4
SciCode61%40%
GSO78.4%4.9%
WeirdML91.9%46.1%
LMArena Coding15191414
ALE-Bench2,041655.35
DeepSWE69.9%—
FrontierCode53.5%—
SWE-bench Verified (bash only)—64.9%
Aider Polyglot—61.3%
LMArena WebDev1625—
FrontierSWE47%—
MirrorCode63.9%—

Agentic & Tool Use Claude Fable 5 leads

Claude Fable 5: 54.0 (#2), Claude Sonnet 4: 38.5 (#31)

BenchmarkClaude Fable 5Claude Sonnet 4
APEX-Agents63.6%—
Remote Labor Index16.1%—
TheAgentCompany—33.1%
τ²-bench Banking39.7%—
Cybench—35%
DeepResearch Bench—46.6%
OSWorld—43.9%
PostTrainBench41.8%—
GBAEval74.5%—
GDP.pdf30%—
LMArena Search1230—
METR Time Horizons—62%
Vending-Bench 25,680—

Reasoning Claude Fable 5 leads

Claude Fable 5: 76.8 (#6), Claude Sonnet 4: 22.9 (#187)

BenchmarkClaude Fable 5Claude Sonnet 4
ARC-AGI-289.2%5.9%
SimpleBench81.9%45.5%
Kagi LLM Benchmark91.4%73%
ARC-AGI-198.5%40%
CritPt28.6%0.3%
EnigmaEval39.3%3.1%
LMArena Hard Prompts15081372
DTBench98.4%77.1%
LMCA61.1%29%
Epoch Capabilities Index162.06141.69
NYT Connections (extended)92.7%—
Chess Puzzles41%—
EBR-Bench39.5%—
Mystery Game Puzzles52%—
Surface Evolver Bench95%—
Bench to the Future 30.13—
ForecastBench—60.2

Math Claude Fable 5 leads

Claude Fable 5: 88.5 (#5), Claude Sonnet 4: 43.3 (#80)

BenchmarkClaude Fable 5Claude Sonnet 4
OTIS Mock AIME 2024-2025100%71.1%
LMArena Math15191375
FrontierMath (Tiers 1-3)87%—
FrontierMath Tier 490.2%—
ProofBench95%—
Omni-MATH—60.2%
MATH Level 5—84.4%
FrontierMath (Feb 2025 set)—4.1%
FrontierMath Erdős0%—
FrontierMath Tier 4 (v1)—0%

Knowledge Claude Fable 5 leads

Claude Fable 5: 62.2 (#25), Claude Sonnet 4: 41.8 (#108)

BenchmarkClaude Fable 5Claude Sonnet 4
GPQA Diamond85.9%79.2%
LMArena Expert15341372
Humanity's Last Exam—7.8%
SimpleQA Verified70.7%—
MMLU-Pro—84.3%
Confabulations—13.2%
Vectara Hallucination Rate—10.3%
GPQA (HELM)—70.6%

Multimodal Claude Fable 5 leads

Claude Fable 5: 45.3 (#17), Claude Sonnet 4: 26.2 (#121)

BenchmarkClaude Fable 5Claude Sonnet 4
LMArena Vision13241191
GeoBench—37%
VPCT—34%
Blueprint-Bench 238.6%—
Furniture Assembly35.8%—
LMArena Document1496—
MindCube—44.8%

Multilingual Claude Fable 5 leads

Claude Fable 5: 57.3 (#9), Claude Sonnet 4: 46.7 (#156)

BenchmarkClaude Fable 5Claude Sonnet 4
LMArena Non-English14811333
LMArena Chinese15431350
LMArena French15051363
LMArena German14861331
LMArena Japanese15061302
LMArena Korean14881291
LMArena Russian15041355
LMArena Spanish14981357

Instruction Following Claude Fable 5 leads

Claude Fable 5: 78.6 (#8), Claude Sonnet 4: 71.7 (#145)

BenchmarkClaude Fable 5Claude Sonnet 4
LMArena Instruction Following15021376
IFEval—84%

Long Context Claude Fable 5 leads

Claude Fable 5: 46.3 (#23), Claude Sonnet 4: 33.7 (#259)

BenchmarkClaude Fable 5Claude Sonnet 4
LMArena Longer Query15091398
Fiction.LiveBench—46.9%

Writing & Preference Claude Fable 5 leads

Claude Fable 5: 75.9 (#5), Claude Sonnet 4: 57.1 (#132)

BenchmarkClaude Fable 5Claude Sonnet 4
LMArena Text14911351
LMArena Creative Writing14941345
EQ-Bench Creative Writing19431483
LMArena Multi-Turn15041376
Short-Story Creative Writing—81.4%
WildBench—83.8%
EQ-Bench 41340—

Frequently asked questions

Is Claude Fable 5 better than Claude Sonnet 4?

Claude Fable 5 is the stronger model overall, scoring 66.8 to 40.8 on the Noometry Index. Claude Sonnet 4 costs 3.3× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Which is cheaper, Claude Fable 5 or Claude Sonnet 4?

Claude Sonnet 4 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; Claude Fable 5 lists at $10 and $50.

Is Claude Fable 5 or Claude Sonnet 4 better for coding?

Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 43.5 in the Noometry coding category.

Which has the bigger context window?

Claude Fable 5 does, with 1M tokens against 200K.

How many benchmarks do Claude Fable 5 and Claude Sonnet 4 share?

34 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and Claude Sonnet 4 has 58.

Related comparisons

Go deeper