Noometry

Model comparison

Claude Fable 5 vs Claude Opus 4.6

Claude Fable 5 is the stronger model overall, scoring 66.8 to 58.2 on the Noometry Index. Claude Opus 4.6 costs 2.0× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

Last verified . 51 shared benchmarks.

Claude Fable 5 Anthropic

66.8

Rank #5 Confirmed

Claude Opus 4.6 Anthropic

58.2

Rank #20 Confirmed

Summary

  • They share 51 benchmarks with published results for both. Claude Fable 5 scores higher in 7 categories and Claude Opus 4.6 in 3 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where Claude Fable 5 leads 88.5 to 63.0.
  • The biggest single-benchmark swing is FrontierMath Tier 4: 90.2% for Claude Fable 5 and 26.8% for Claude Opus 4.6.
  • Claude Opus 4.6 is cheaper at $5 / $25 per million input/output tokens, against $10 / $50 for Claude Fable 5.

Side by side

Claude Fable 5Claude Opus 4.6
ProviderAnthropicAnthropic
Noometry Index66.858.2
Released2026-06-072026-02-04
WeightsProprietaryProprietary
Context window1M1M
Max output128K128K
Input $ / M tokens$10$5
Output $ / M tokens$50$25
Results tracked6268

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 Opus 4.6: 57.2 (#20)

BenchmarkClaude Fable 5Claude Opus 4.6
FrontierCode53.5%26.6%
LMArena WebDev16251547
GSO78.4%41.2%
WeirdML91.9%78%
LMArena Coding15191536
ALE-Bench2,041996.5
SWE-bench Verified—78.7%
DeepSWE69.9%—
SWE-bench Verified (bash only)—75.6%
SWE-bench Multilingual—72%
FrontierSWE47%—
SciCode61%—
MirrorCode63.9%—
AlgoTune—1.47

Agentic & Tool Use Claude Fable 5 leads

Claude Fable 5: 54.0 (#2), Claude Opus 4.6: 51.1 (#4)

BenchmarkClaude Fable 5Claude Opus 4.6
APEX-Agents63.6%46.3%
Remote Labor Index16.1%4.2%
τ²-bench Banking39.7%27.3%
GBAEval74.5%44.1%
LMArena Search12301253
Vending-Bench 25,6808,018
Terminal-Bench—79.8%
Cybench—93%
DeepResearch Bench—55.3%
PostTrainBench41.8%—
GDP.pdf30%—
METR Time Horizons—78.9%

Reasoning Claude Fable 5 leads

Claude Fable 5: 76.8 (#6), Claude Opus 4.6: 57.8 (#23)

BenchmarkClaude Fable 5Claude Opus 4.6
ARC-AGI-289.2%69.2%
SimpleBench81.9%67.6%
Kagi LLM Benchmark91.4%83.6%
NYT Connections (extended)92.7%92.1%
ARC-AGI-198.5%94%
Chess Puzzles41%17%
EnigmaEval39.3%7.6%
EBR-Bench39.5%12.7%
LMArena Hard Prompts15081527
Mystery Game Puzzles52%25%
DTBench98.4%91.2%
LMCA61.1%55.8%
Epoch Capabilities Index162.06155.24
CritPt28.6%—
Thematic Generalization—80.6%
Surface Evolver Bench95%—
Bench to the Future 30.13—
ForecastBench—60

Math Claude Fable 5 leads

Claude Fable 5: 88.5 (#5), Claude Opus 4.6: 63.0 (#31)

BenchmarkClaude Fable 5Claude Opus 4.6
FrontierMath (Tiers 1-3)87%66%
FrontierMath Tier 490.2%26.8%
OTIS Mock AIME 2024-2025100%94.4%
ProofBench95%50%
LMArena Math15191519
MathArena Final-Answer Competitions—78.5%
FrontierMath (Feb 2025 set)—40.7%
FrontierMath Erdős0%—
FrontierMath Tier 4 (v1)—22.9%

Knowledge Too close to call

Claude Fable 5: 62.2 (#25), Claude Opus 4.6: 61.9 (#26)

BenchmarkClaude Fable 5Claude Opus 4.6
GPQA Diamond85.9%90.5%
SimpleQA Verified70.7%47%
LMArena Expert15341546
Humanity's Last Exam—34.4%
Vectara Hallucination Rate—12.2%

Multimodal Claude Fable 5 leads

Claude Fable 5: 45.3 (#17), Claude Opus 4.6: 37.3 (#74)

BenchmarkClaude Fable 5Claude Opus 4.6
LMArena Vision13241316
Furniture Assembly35.8%28.3%
LMArena Document14961507
Blueprint-Bench 238.6%—

Multilingual Too close to call

Claude Fable 5: 57.3 (#9), Claude Opus 4.6: 57.9 (#6)

BenchmarkClaude Fable 5Claude Opus 4.6
LMArena Non-English14811489
LMArena Chinese15431551
LMArena French15051513
LMArena German14861502
LMArena Japanese15061484
LMArena Korean14881464
LMArena Russian15041497
LMArena Spanish14981510

Instruction Following Too close to call

Claude Fable 5: 78.6 (#8), Claude Opus 4.6: 79.5 (#4)

BenchmarkClaude Fable 5Claude Opus 4.6
LMArena Instruction Following15021523

Long Context Claude Opus 4.6 leads

Claude Fable 5: 46.3 (#23), Claude Opus 4.6: 48.1 (#13)

BenchmarkClaude Fable 5Claude Opus 4.6
LMArena Longer Query15091520
CL-bench—20.7%
CL-bench Life—17%

Writing & Preference Claude Fable 5 leads

Claude Fable 5: 75.9 (#5), Claude Opus 4.6: 73.5 (#10)

BenchmarkClaude Fable 5Claude Opus 4.6
LMArena Text14911503
LMArena Creative Writing14941505
EQ-Bench Creative Writing19431809
EQ-Bench 413401223
LMArena Multi-Turn15041513

Frequently asked questions

Is Claude Fable 5 better than Claude Opus 4.6?

Claude Fable 5 is the stronger model overall, scoring 66.8 to 58.2 on the Noometry Index. Claude Opus 4.6 costs 2.0× 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 Opus 4.6?

Claude Opus 4.6 is cheaper. It lists at $5 per million input tokens and $25 per million output tokens; Claude Fable 5 lists at $10 and $50.

Is Claude Fable 5 or Claude Opus 4.6 better for coding?

Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 57.2 in the Noometry coding category.

Which has the bigger context window?

Both accept 1M tokens.

How many benchmarks do Claude Fable 5 and Claude Opus 4.6 share?

51 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and Claude Opus 4.6 has 68.

Related comparisons

Go deeper