Noometry

Anthropic, proprietary

Claude Fable 5

Claude Fable 5 by Anthropic ranks 5th of 354 ranked models on the Noometry Index as of October 2026, with a score of 66.8. Its strongest category is agentic & tool use, where it ranks 2nd. API pricing starts at $10 per million input tokens and $50 per million output tokens, with a 1M-token context window.

Last verified

Noometry rank
#5 of 354
Index score
66.8
Evidence
Confirmed 62 results
Provider
Anthropic
Released
June 7, 2026
Weights
Proprietary
Reasoning
Yes
Context window
1M
Max output
128K
Input price
$10 / M
Output price
$50 / M
Blended price
$20 / M
Output speed
25 tokens/s Kagi
Value
#208 of 219
Knowledge cutoff
January 2026
Input
text, image, pdf

Category scores

Each category score combines every public result we have in that category.

  1. Coding 70.6
  2. Agentic & Tool Use 54.0
  3. Reasoning 76.8
  4. Math 88.5
  5. Knowledge 62.2
  6. Multimodal 45.3
  7. Multilingual 57.3
  8. Instruction Following 78.6
  9. Long Context 46.3
  10. Writing & Preference 75.9
CategoryScoreRankResults
Coding70.6#49
Agentic & Tool Use54.0#26
Reasoning76.8#614
Math88.5#55
Knowledge62.2#253
Multimodal45.3#173
Multilingual57.3#91
Instruction Following78.6#81
Long Context46.3#231
Writing & Preference75.9#55

Strengths and weaknesses

Categories where Claude Fable 5 places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Coding70.6+31.9#4 of 340, top 2%
Agentic & Tool Use54.0+23.7#2 of 154, top 2%
Math88.5+51.9#5 of 327, top 2%

Weakest categories

CategoryScorevs medianRank
Multimodal45.3+6.7#17 of 128, top 14%
Knowledge62.2+24.9#25 of 314, top 8%
Long Context46.3+5.4#23 of 296, top 8%

Closest competitors

The models ranked just above and below Claude Fable 5. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
GPT-6 Astra#170.8$20—Compare
Claude Fable 5.1#269.0$20—Compare
Claude Opus 5.5#368.6$8—Compare
Claude Opus 5#467.8$10—Compare
GPT-6.1 Sol#665.6$4—Compare
GPT-5.6 Sol#765.0$810Compare
GPT-5.5 Pro#864.3$67.50—Compare
GPT-5.5#963.4$11.2525Compare

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

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
DeepSWE68.6%highEpoch AI
DeepSWE59.6%lowEpoch AI
DeepSWE69.7%maxEpoch AI
DeepSWE65.4%mediumEpoch AI
DeepSWE69.9%#6 of 29, top 21%xhighEpoch AI
FrontierCode53.5%#2 of 37, top 6%Epoch AI
LMArena WebDev1625#16 of 113, top 15%highLMArena2026-10-08
FrontierSWE47%#7 of 18, top 39%maxEpoch AI
SciCode61%#4 of 121, top 4%maxEpoch AI
GSO78.4%#3 of 31, top 10%Epoch AI
WeirdML87.8%highEpoch AI
WeirdML91.9%#3 of 119, top 3%maxEpoch AI
LMArena Coding1519#7 of 294, top 3%highLMArena2026-10-08
MirrorCode63.9%#3 of 9, top 34%highEpoch AI2026-08-10
ALE-Bench2,041#7 of 105, top 7%highEpoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
APEX-Agents63.6%#10 of 49, top 21%Epoch AI
Remote Labor Index16.1%#3 of 14, top 22%Epoch AI
τ²-bench Banking39.7%#8 of 26, top 31%maxτ²-bench2026-08-04
PostTrainBench41.8%Best of 11maxEpoch AI
GBAEval74.5%#2 of 23, top 9%Epoch AI
GDP.pdf30%#5 of 36, top 14%Epoch AI
GDP.pdf29.8%maxEpoch AI
LMArena Search1230#5 of 32, top 16%LMArena2026-08-24
Vending-Bench 24,530Epoch AI
Vending-Bench 25,680#22 of 60, top 37%highEpoch AI
Vending-Bench 25,019lowEpoch AI
Vending-Bench 24,967maxEpoch AI
Vending-Bench 24,340mediumEpoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
ARC-AGI-287.5%highEpoch AI
ARC-AGI-276.8%lowEpoch AI
ARC-AGI-289.2%#8 of 83, top 10%maxEpoch AI
ARC-AGI-282.5%mediumEpoch AI
ARC-AGI-288.3%xhighEpoch AI
SimpleBench81.9%Best of 77maxEpoch AI
Kagi LLM Benchmark88.8%Kagi LLM Benchmark
Kagi LLM Benchmark91.4%Best of 99Kagi LLM Benchmark
NYT Connections (extended)92.7%#11 of 91, top 13%high reasoningLech Mazur benchmarks
ARC-AGI-195.5%highEpoch AI
ARC-AGI-190.5%lowEpoch AI
ARC-AGI-198.5%Best of 83maxEpoch AI
ARC-AGI-192.5%mediumEpoch AI
ARC-AGI-198.5%xhighEpoch AI
CritPt28.6%#12 of 134, top 9%maxEpoch AI
Chess Puzzles41%#18 of 129, top 14%highEpoch AI2026-08-06
Chess Puzzles28%lowEpoch AI2026-08-06
Chess Puzzles41%maxEpoch AI2026-06-09
EnigmaEval39.3%Best of 38highEpoch AI
EBR-Bench39.5%#8 of 24, top 34%maxEpoch AI2026-07-17
LMArena Hard Prompts1508#8 of 297, top 3%highLMArena2026-10-08
Mystery Game Puzzles52%#10 of 74, top 14%maxEpoch AI2026-07-31
DTBench96.5%highEpoch AI
DTBench91.5%lowEpoch AI
DTBench98.4%#2 of 151, top 2%maxEpoch AI
DTBench95.7%mediumEpoch AI
DTBench97.3%xhighEpoch AI
LMCA61.1%#5 of 125, top 4%highEpoch AI
LMCA59.6%lowEpoch AI
LMCA60.3%maxEpoch AI
LMCA60.3%mediumEpoch AI
LMCA61%xhighEpoch AI
Surface Evolver Bench95%Best of 25highEpoch AI
Bench to the Future 30.13#9 of 10, top 90%highEpoch AI
Epoch Capabilities Index162.06#9 of 213, top 5%Epoch AI2026-06-09

Math

BenchmarkScorePositionSettingSourceDate
FrontierMath (Tiers 1-3)87%#9 of 81, top 12%maxEpoch AI2026-06-09
FrontierMath Tier 490.2%#4 of 63, top 7%maxEpoch AI2026-06-09
OTIS Mock AIME 2024-2025100%Best of 173highEpoch AI2026-08-06
OTIS Mock AIME 2024-202597.8%lowEpoch AI2026-08-06
OTIS Mock AIME 2024-202599.7%maxEpoch AI2026-06-10
ProofBench95%#8 of 77, top 11%maxEpoch AI
LMArena Math1519#5 of 285, top 2%highLMArena2026-10-08
FrontierMath Erdős0%#4 of 7, top 58%maxEpoch AI2026-08-28

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond83.3%highEpoch AI2026-08-06
GPQA Diamond78.8%lowEpoch AI2026-08-06
GPQA Diamond85.9%#61 of 186, top 33%maxEpoch AI2026-08-06
SimpleQA Verified70.7%#6 of 77, top 8%xhighEpoch AI2026-08-10
LMArena Expert1534#8 of 273, top 3%highLMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1324Best of 122highLMArena2026-10-09
Blueprint-Bench 238.6%#5 of 31, top 17%Epoch AI
Furniture Assembly35.8%#18 of 31, top 59%maxEpoch AI2026-09-10
LMArena Document1496#4 of 38, top 11%LMArena2026-09-13

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1481#9 of 297, top 4%highLMArena2026-10-08
LMArena Chinese1543#8 of 285, top 3%highLMArena2026-10-08
LMArena French1505#8 of 223, top 4%highLMArena2026-10-08
LMArena German1486#17 of 231, top 8%highLMArena2026-10-08
LMArena Japanese1506#5 of 211, top 3%highLMArena2026-10-08
LMArena Korean1488#4 of 213, top 2%highLMArena2026-10-08
LMArena Russian1504#7 of 283, top 3%highLMArena2026-10-08
LMArena Spanish1498#7 of 226, top 4%highLMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1502#6 of 298, top 3%highLMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1509#6 of 291, top 3%highLMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1491#8 of 297, top 3%highLMArena2026-10-08
LMArena Creative Writing1494#5 of 295, top 2%highLMArena2026-10-08
EQ-Bench Creative Writing1943#10 of 115, top 9%EQ-Bench
EQ-Bench 41340#2 of 28, top 8%EQ-Bench
LMArena Multi-Turn1504#4 of 295, top 2%highLMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
anthropic$10$50$12026-10-10
azure$10$50$12026-10-10
bedrock$10$50$12026-10-10
openrouter$10$50$12026-10-10
vertex$10$50$12026-10-10

All Anthropic API prices → Estimate your cost →

Compare Claude Fable 5

Other Anthropic models

Frequently asked questions

How good is Claude Fable 5?

Claude Fable 5 by Anthropic ranks 5th of 354 ranked models on the Noometry Index as of October 2026, with a score of 66.8. Its strongest category is agentic & tool use, where it ranks 2nd. API pricing starts at $10 per million input tokens and $50 per million output tokens, with a 1M-token context window.

How much does Claude Fable 5 cost?

Claude Fable 5 costs $10 per million input tokens and $50 per million output tokens on Anthropic's own API, with cached input at $1.

What is Claude Fable 5's context window?

Claude Fable 5 accepts up to 1M tokens of input and can write up to 128K tokens in one response.

Is Claude Fable 5 open source?

No. Claude Fable 5 is proprietary and available only through Anthropic's API and partner platforms.

How fast is Claude Fable 5?

Claude Fable 5 generated about 25 output tokens per second in the Kagi LLM Benchmark's timed runs. Speed varies by provider, load and reasoning effort.

What are Claude Fable 5's strengths and weaknesses?

Relative to other ranked models, Claude Fable 5 places best in coding, agentic & tool use, math and lowest in multimodal, knowledge, long context.

What is Claude Fable 5 best at?

Its best category is agentic & tool use, where it ranks 2nd on Noometry.