Language modelActive

Claude Fable 5

Anthropic

Released: June 9, 2026
Data date: September 8, 2026
Predecessor: Claude Mythos Preview (Official documentation)
Successor: Claude Fable 5.1

Benchmark-Meta-Score: 85.1 / 100
Provisional, 59.2 % panel coverage
Speed: 61.7 tokens/s
Output throughput
Price: $10 · $50
Input · Output / 1M tokens
Input: Text, Image
Output: Text

Claude Fable 5 was the generally available half of the model core it shared with Claude Mythos 5 from June 9, 2026. It was the direct predecessor to Claude Fable 5.1, combining demanding software work, knowledge work, and vision with safety classifiers.

Its availability followed an unusual path. A report from Amazon researchers described a method that could bypass Fable 5 safeguards during vulnerability research, after which US export controls suspended access from June 12 through June 30. Fable returned globally on July 1. Anthropic now labels it legacy, while keeping its API available and pointing new integrations to Fable 5.1.

Specifications and access

Specification	Value and source
Weights	ProprietarySource
Access	APISource
Context window	1,000,000 tokensSource
Knowledge cutoff	January 2026Source
Notes	Mythos-class tier above Opus, most capable widely-released model, 1M context (new tokenizer, ~30% more tokens per text vs. pre-Opus-4.7 models), 128K output, adaptive thinking only, safety classifiers with refusal fallback to Opus 4.8. Briefly suspended 2026-06-12 to 2026-06-30 under a US export control directive after a jailbreak was found; generally available again since 2026-07-01.Source
API model ID	claude-fable-5Source
Source status	active (legacy)Source
Release	2026-06-09Source
Maximum output	128,000 tokensSource
Input	Text, imagesSource
Output	TextSource
Reasoning mode	adaptive (always on); default highSource
Knowledge cutoff	2026-01Source
Tool calling	YesSource
Structured output	YesSource

Benchmark meta-analysis

Overall and area scores

These Gradually indexes place selected published measurements on a 0 to 100 scale. They are not accuracy percentages and do not guarantee results in your application.

Retrieval dates of the sources used 2026-08-26 - 2026-10-19

Overall index

Index 85.1 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 59 %
Test families: 43 / 71
Sources: 5
Mathematical bounds with missing values: 50.3-91.2 / 100

Coding

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 46 %
Test families: 8 / 15
Sources: 3
Mathematical bounds with missing values: 39.1-93.5 / 100

Reasoning

Index 79.1 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 79 %
Test families: 6 / 7
Sources: 4
Mathematical bounds with missing values: 62.1-83.5 / 100

Mathematics

Index 88.6 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 60 %
Test families: 3 / 5
Sources: 2
Mathematical bounds with missing values: 53.2-93.2 / 100

Knowledge

Index 98.1 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 67 %
Test families: 2 / 3
Sources: 2
Mathematical bounds with missing values: 65.4-98.7 / 100

Agents

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 33 %
Test families: 4 / 12
Sources: 1
Mathematical bounds with missing values: 26.9-93.5 / 100

General

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 40 %
Test families: 2 / 5
Sources: 2
Mathematical bounds with missing values: 28.7-88.7 / 100

Finance

Index 93.1 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 100 %
Test families: 7 / 7
Sources: 2
Mathematical bounds with missing values: 93.1-93.1 / 100

Legal

Index 94.4 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 100 %
Test families: 4 / 4
Sources: 2
Mathematical bounds with missing values: 94.4-94.4 / 100

Health

Index 91.9 / 100

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 67 %
Test families: 2 / 3
Sources: 1
Mathematical bounds with missing values: 61.3-94.6 / 100

Multimodal

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 38 %
Test families: 2 / 4
Sources: 2
Mathematical bounds with missing values: 37.0-99.5 / 100

Search

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 100 %
Test families: 1 / 1
Sources: 1
Mathematical bounds with missing values: 53.8-53.8 / 100

Productivity

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 50 %
Test families: 1 / 2
Sources: 1
Mathematical bounds with missing values: 44.7-94.7 / 100

Education

Not enough coverage for a score yet

provisional

Published configurations differ within this comparison cohort. Treat the index as provisional.

Panel-weight coverage: 50 %
Test families: 1 / 2
Sources: 1
Mathematical bounds with missing values: 44.8-94.8 / 100

Documented method · Included test families and measurements

Method version: benchmark-meta-score-v1

The index includes active measurements from reviewed primary sources when at least five models are compared in the same documented comparison slice across dataset, harness, agent, tools, and trial count. Each benchmark series and evaluator contributes one shared setup slice from the newest capture; when dates match, the wider slice is used. When one source and series include several explicitly documented reasoning tiers, the highest documented tier (none through max) is selected, never the best result value. The explicit source label “Non-reasoning” is treated as none. When selecting between multiple results, unknown or ambiguous tiers exclude the model from that series. Unambiguous single configurations do not require a tier ordering. The best individual result for each model is not selected. Equal raw values share the same rank percentile. Multiple series from the same evaluator within a family receive equal weight; evaluators within a family, families within an area, and areas in the overall index then receive equal weight. Unmeasured series remain visible in coverage and bounds. An overall score requires at least four areas, six test families, and two sources. An area score requires at least two test families. Every overall and area score also requires at least 50% of its panel weight.

Published model-specific parameters can differ. This is not a controlled experiment; comparisons with such differences are marked provisional.

Unmeasured portions are not entered as zero. They reduce coverage and contribute only to the displayed bounds. Scores with different coverage are therefore not equally complete.

The bounds hold observed percentiles fixed and allow missing contributions anywhere from 0 to 100. They are not confidence intervals or predictions of future recalculations.

The source count is not a count of independent experiments.

OECD handbook on composite indicators

All 43 test families and 12 of 47 raw measurements are listed here. The complete calculation for every model is available in the JSON export.

Test family: AIME
Mathematics · Family index: 90.9 / 100
Nominal family weight in the overall panel: 1.43 % · Coverage of this test family: 100 %
- Epoch AI benchmark run dataset snapshot
  Published configurations differ within this cohort.
  Measurement:
  epoch-otis-mock-aime-2024-2025
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  1.43 %
  Raw value:
  1 ratio
  Metric:
  model_extracted_exact_match
  Version:
  1.1.6
  Participating models:
  23
  Rank percentile within the cohort:
  90.9 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  epoch-score-design-v1:["OTIS Mock AIME 2024-2025","OTIS Mock AIME 2024-2025","1.1.6","model_extracted_exact_match"]
  Reasoning tier:
  high
  Documented comparison setup
  Publication cohort:
  epoch-score-design-v1:["OTIS Mock AIME 2024-2025","OTIS Mock AIME 2024-2025","1.1.6","model_extracted_exact_match"]
  Dataset type:
  not-reported-in-captured-run-csv
  Setup and model parameters
  Published reasoning tier: high. The capture does not report other inference parameters or a compute budget.
  Independent evaluation. Epoch run alias: claude-fable-5_high; Epoch model label: Claude Fable 5; Display label field: Claude Fable 5 (high); Reasoning setting encoded in the alias: high. Task ID: OTIS Mock AIME 2024-2025. Task version: 1.1.6. Run ID: hwrnvtEbDAocFtjcsHqsQh. Started at 2026-08-06T22:18:19.000Z. Original task label: OTIS Mock AIME 2024-2025. The published row reports mean_score=1.0 and stderr=0.0. The CSV has no sample-count column, so none is inferred. Standard error is not presented as a confidence interval. Meta-score selection: Per model, the highest documented reasoning tier is selected within the same Epoch source test design, original task, version, and scorer, independently of the measured value. This derived score copy does not claim matched compute budgets; the capture publishes neither a common budget nor all inference parameters. The raw model-profile observation remains unchanged. Selected reasoning tier: high.
Test family: AA-AnalystAgent
Agents · Family index: 73.7 / 100
Nominal family weight in the overall panel: 0.6 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-aa-analyst-agent
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  0.6 %
  Raw value:
  48.75 %
  Metric:
  pass^5 reliability
  Version:
  AA cohort: 80 questions
  Participating models:
  39
  Rank percentile within the cohort:
  73.7 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-aa-analyst-agent:AA cohort: 80 questions:standalone:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-aa-analyst-agent","AA cohort: 80 questions","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Harness:
  Artificial Analysis Stirrup
  Agent:
  Stirrup analyst agent
  Tool policy:
  100-turn cap; Python sandbox with source files, URL fetch, image viewing, and answer submission
  Attempts:
  pass^5, 5 independent attempts per question
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption. 80 quantitative questions across 14 domains are run five independent times. pass^5 counts only questions answered correctly in all five attempts.. Harness: Artificial Analysis Stirrup. Agent: Stirrup analyst agent. Tool policy: 100-turn cap; Python sandbox with source files, URL fetch, image viewing, and answer submission. Attempts: pass^5, 5 independent attempts per question. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption. 80 quantitative questions across 14 domains are run five independent times. pass^5 counts only questions answered correctly in all five attempts.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Harness: Artificial Analysis Stirrup. Agent: Stirrup analyst agent. Tool policy: 100-turn cap; Python sandbox with source files, URL fetch, image viewing, and answer submission. Attempts: pass^5, 5 independent attempts per question. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Score display: 48.8%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: AA-Briefcase
Agents · Family index: 87.8 / 100
Nominal family weight in the overall panel: 0.6 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-aa-briefcase
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  0.6 %
  Raw value:
  1,539.25 Elo
  Metric:
  AA-Briefcase Elo
  Version:
  v1.1
  Participating models:
  83
  Rank percentile within the cohort:
  87.8 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-aa-briefcase:v1.1:v4.3.2:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-aa-briefcase","v1.1","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: AA-LCR
Reasoning · Family index: 84.1 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-aa-lcr
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  1.02 %
  Raw value:
  82.3333333333333 %
  Metric:
  Pass rate
  Version:
  v1.1
  Participating models:
  83
  Rank percentile within the cohort:
  84.1 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-aa-lcr:v1.1:v4.3.2:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-aa-lcr","v1.1","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Score display: 82.3%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: AutomationBench-AA
Agents · Family index: 73.2 / 100
Nominal family weight in the overall panel: 0.6 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-automationbench-aa
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  0.6 %
  Raw value:
  54.071301140653816 %
  Metric:
  Objectives completed without guardrail violations
  Version:
  unspecified
  Participating models:
  83
  Rank percentile within the cohort:
  73.2 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-automationbench-aa:unspecified:v4.3.2:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-automationbench-aa","unspecified","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Score display: 54.1%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: GDP.pdf
Reasoning · Family index: 90.2 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-gdp-pdf
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  1.02 %
  Raw value:
  24 %
  Metric:
  All-pass rate
  Version:
  unspecified
  Participating models:
  83
  Rank percentile within the cohort:
  90.2 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-gdp-pdf:unspecified:v4.3.2:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-gdp-pdf","unspecified","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Score display: 24.0%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: GDPval-AA
Agents · Family index: 87.8 / 100
Nominal family weight in the overall panel: 0.6 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-gdpval-aa
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  0.6 %
  Raw value:
  1,609.83 Elo
  Metric:
  GDPval-AA Elo
  Version:
  v2.1
  Participating models:
  83
  Rank percentile within the cohort:
  87.8 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-gdpval-aa:v2.1:v4.3.2:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-gdpval-aa","v2.1","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: IFBench
General · Family index: 49.3 / 100
Nominal family weight in the overall panel: 1.43 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-ifbench
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  1.43 %
  Raw value:
  63.469387755102005 %
  Metric:
  IFBench prompt-level loose accuracy (pass@1)
  Version:
  AA test set: 294 questions
  Participating models:
  70
  Rank percentile within the cohort:
  49.3 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-ifbench:AA test set: 294 questions:standalone:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-ifbench","AA test set: 294 questions","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Tool policy:
  Single-turn answer evaluated with the official IFBench source code
  Attempts:
  pass@1, 5 repeats per question
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption. Artificial Analysis runs 294 questions from the IFBench test set with five repeats per question. The official evaluator scores prompt-level accuracy in loose mode as pass@1.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Tool policy: Single-turn answer evaluated with the official IFBench source code. Attempts: pass@1, 5 repeats per question. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Score display: 63.5%. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: AA-Omniscience
Knowledge · Family index: 98.8 / 100
Nominal family weight in the overall panel: 2.38 % · Coverage of this test family: 100 %
- Artificial Analysis
  Published configurations differ within this cohort.
  Measurement:
  artificial-analysis-omniscience
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  2.38 %
  Raw value:
  43.3 points
  Metric:
  AA-Omniscience Index
  Version:
  unspecified
  Participating models:
  83
  Rank percentile within the cohort:
  98.8 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  artificial-analysis-omniscience:unspecified:v4.3.2:reasoning:tokens=0:suffix=Max%2C%20Opus%204.8%20Fallback
  Reasoning tier:
  Max, Opus 4.8 Fallback
  Documented comparison setup
  Publication cohort:
  ["artificial-analysis-omniscience","unspecified","aa-model-page-2026-10-19-8a022369738c"]
  Dataset type:
  artificial-analysis-evaluation
  Setup and model parameters
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback
  Exact AA configuration: Claude Fable 5 (Max, Opus 4.8 Fallback) (release claude-fable-5, row claude-fable-5, UUID cd55210d-358e-4df1-ba9c-9acb5f186cc9, reasoning=true, source reasoning-token configuration budget=0, suffix=Max, Opus 4.8 Fallback). This configured value is not measured token consumption.. Source model ID: artificial-analysis/cd55210d-358e-4df1-ba9c-9acb5f186cc9/claude-fable-5/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: Claude Fable 5 (Max, Opus 4.8 Fallback). Evaluator: Artificial Analysis. Dataset type: artificial-analysis-evaluation. Reasoning: reasoning enabled. Reasoning effort: Max, Opus 4.8 Fallback. Source ID: artificial-analysis. Evidence capture ID: aa-model-page-2026-10-19-8a022369738c. Source payload hash: 8a022369738cbe2a0e2f9bf486a2882865dbb0c6f1396dfec84059a802560e3c. Parser version: artificial-analysis-rsc-models-v2. Imported at: 2026-10-19
Test family: EnigmaEval
Reasoning · Family index: 100.0 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
- Scale AI
  Published configurations differ within this cohort.
  Measurement:
  enigma-eval
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  1.02 %
  Raw value:
  39.28 %
  Metric:
  pass@1 accuracy
  Version:
  open-source release 2026-07-23
  Participating models:
  5
  Rank percentile within the cohort:
  100.0 %
  Retrieved:
  2026-08-26
  Comparison cohort:
  scale-enigma-eval-2026-07-23
  Reasoning tier:
  high
  Documented comparison setup
  Publication cohort:
  scale-enigma-eval-2026-07-23
  Dataset type:
  public
  Attempts:
  pass@1
  Setup and model parameters
  Complex multimodal puzzles with exact answer matching and one attempt.. Attempts: pass@1. Reasoning effort: high
  Complex multimodal puzzles with exact answer matching and one attempt.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-high. Evaluator: Scale AI. Dataset type: public. Leaderboard date: 2026-07-23. Attempts: pass@1. Reasoning effort: high. Score display: 39.28% ± 2.8. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-de06ddcfa8a8. Source payload hash: de06ddcfa8a81a2242bdcf8cae4bc1d2e0d1d9e896cb450a19ce0227b96075a7. Parser version: curated-v1. Imported at: 2026-10-15
Test family: Chess Puzzles
Reasoning · Family index: 61.9 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
- Epoch AI benchmark run dataset snapshot
  Published configurations differ within this cohort.
  Measurement:
  epoch-chess-puzzles
  Series weight within the test family:
  100 %
  Nominal contribution to the overall panel:
  1.02 %
  Raw value:
  0.47 ratio
  Metric:
  exact
  Version:
  2.0.0
  Participating models:
  22
  Rank percentile within the cohort:
  61.9 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  epoch-score-design-v1:["Chess Puzzles","Chess Puzzles","2.0.0","exact"]
  Reasoning tier:
  max
  Documented comparison setup
  Publication cohort:
  epoch-score-design-v1:["Chess Puzzles","Chess Puzzles","2.0.0","exact"]
  Dataset type:
  not-reported-in-captured-run-csv
  Setup and model parameters
  Published reasoning tier: max. The capture does not report other inference parameters or a compute budget.
  Independent evaluation. Epoch run alias: claude-fable-5_max; Epoch model label: Claude Fable 5; Display label field: Claude Fable 5 (max); Reasoning setting encoded in the alias: max. Task ID: Chess Puzzles. Task version: 2.0.0. Run ID: bxS76CJuVws8iwWU2Wr2PM. Started at 2026-10-13T13:34:37.000Z. Original task label: Chess Puzzles. The published row reports mean_score=0.47 and stderr=0.05016135580465918. The CSV has no sample-count column, so none is inferred. Standard error is not presented as a confidence interval. Meta-score selection: Per model, the highest documented reasoning tier is selected within the same Epoch source test design, original task, version, and scorer, independently of the measured value. This derived score copy does not claim matched compute budgets; the capture publishes neither a common budget nor all inference parameters. The raw model-profile observation remains unchanged. Selected reasoning tier: max.
Test family: FrontierMath
Mathematics · Family index: 90.4 / 100
Nominal family weight in the overall panel: 1.43 % · Coverage of this test family: 100 %
- Epoch AI benchmark run dataset snapshot
  Published configurations differ within this cohort.
  Measurement:
  epoch-frontiermath-tier-4-v2-private
  Series weight within the test family:
  50 %
  Nominal contribution to the overall panel:
  0.71 %
  Raw value:
  0.902 ratio
  Metric:
  verification_code
  Version:
  2.0.0
  Participating models:
  19
  Rank percentile within the cohort:
  94.4 %
  Retrieved:
  2026-10-19
  Comparison cohort:
  epoch-score-design-v1:["FrontierMath-Tier-4-v2-Private","FrontierMath-Tier-4-v2-Private","2.0.0","verification_code"]
  Reasoning tier:
  max
  Documented comparison setup
  Publication cohort:
  epoch-score-design-v1:["FrontierMath-Tier-4-v2-Private","FrontierMath-Tier-4-v2-Private","2.0.0","verification_code"]
  Dataset type:
  not-reported-in-captured-run-csv
  Setup and model parameters
  Published reasoning tier: max. The capture does not report other inference parameters or a compute budget.
  Independent evaluation. Epoch run alias: claude-fable-5_max; Epoch model label: Claude Fable 5; Display label field: Claude Fable 5 (max); Reasoning setting encoded in the alias: max. Task ID: FrontierMath-Tier-4-v2-Private. Task version: 2.0.0. Run ID: ReDAtJGGfRtrEGGWf4Zbsd. Started at 2026-06-09T18:12:35.000Z. Original task label: FrontierMath-Tier-4-v2-Private. The published row reports mean_score=0.902 and stderr=0.046. The CSV has no sample-count column, so none is inferred. Standard error is not presented as a confidence interval. Meta-score selection: Per model, the highest documented reasoning tier is selected within the same Epoch source test design, original task, version, and scorer, independently of the measured value. This derived score copy does not claim matched compute budgets; the capture publishes neither a common budget nor all inference parameters. The raw model-profile observation remains unchanged. Selected reasoning tier: max.
Test family: GPQA Diamond
Reasoning · Family index: 48.5 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: Humanity’s Last Exam
Reasoning · Family index: 100.0 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 50 %
Test family: LMArena Code Arena WebDev
Coding · Family index: 66.7 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 100 %
Test family: LMArena Search
Search · Family index: 53.8 / 100
Nominal family weight in the overall panel: 7.14 % · Coverage of this test family: 100 %
Test family: LMArena Text, Style Control
General · Family index: 94.4 / 100
Nominal family weight in the overall panel: 1.43 % · Coverage of this test family: 100 %
Test family: LMArena Vision, Style Control
Multimodal · Family index: 100.0 / 100
Nominal family weight in the overall panel: 1.79 % · Coverage of this test family: 100 %
Test family: MMMU-Pro
Multimodal · Family index: 95.7 / 100
Nominal family weight in the overall panel: 1.79 % · Coverage of this test family: 50 %
Test family: PRBench Finance
Finance · Family index: 100.0 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: PRBench Legal
Legal · Family index: 100.0 / 100
Nominal family weight in the overall panel: 1.79 % · Coverage of this test family: 100 %
Test family: SWE-bench Verified
Coding · Family index: 91.5 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 50 %
Test family: Terminal-Bench
Coding · Family index: 87.5 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 83 %
Test family: Terminal-Bench Science
Coding · Family index: 67.7 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 50 %
Test family: Code Migration
Coding · Family index: 87.1 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 100 %
Test family: CorpFin v2
Finance · Family index: 98.6 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: EMB
Finance · Family index: 94.9 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: Finance Agent (v2)
Finance · Family index: 81.7 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: LiveCodeBench v6
Coding · Family index: 98.7 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 100 %
Test family: Harvey's Legal Agent Benchmark
Legal · Family index: 86.1 / 100
Nominal family weight in the overall panel: 1.79 % · Coverage of this test family: 100 %
Test family: LegalBench
Legal · Family index: 100.0 / 100
Nominal family weight in the overall panel: 1.79 % · Coverage of this test family: 100 %
Test family: Legal Research Bench
Legal · Family index: 91.7 / 100
Nominal family weight in the overall panel: 1.79 % · Coverage of this test family: 100 %
Test family: MedCode
Health · Family index: 97.0 / 100
Nominal family weight in the overall panel: 2.38 % · Coverage of this test family: 100 %
Test family: MedScribe
Health · Family index: 86.8 / 100
Nominal family weight in the overall panel: 2.38 % · Coverage of this test family: 100 %
Test family: MMLU-Pro
Knowledge · Family index: 97.4 / 100
Nominal family weight in the overall panel: 2.38 % · Coverage of this test family: 100 %
Test family: MortgageTax
Finance · Family index: 88.5 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: ProgramBench
Coding · Family index: 83.7 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 100 %
Test family: ProofBench
Mathematics · Family index: 84.6 / 100
Nominal family weight in the overall panel: 1.43 % · Coverage of this test family: 100 %
Test family: SAGE
Education · Family index: 89.7 / 100
Nominal family weight in the overall panel: 3.57 % · Coverage of this test family: 100 %
Test family: Public Benefits Bench v1.1
Productivity · Family index: 89.5 / 100
Nominal family weight in the overall panel: 3.57 % · Coverage of this test family: 100 %
Test family: Tax Agent Bench
Finance · Family index: 89.1 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: TaxEval v2
Finance · Family index: 98.7 / 100
Nominal family weight in the overall panel: 1.02 % · Coverage of this test family: 100 %
Test family: Vibe Code Bench v1.1
Coding · Family index: 97.2 / 100
Nominal family weight in the overall panel: 0.48 % · Coverage of this test family: 100 %

Benchmark sources at a glance

Each evaluator measures different capabilities under its own conditions. Test series, composite indices and their components stay separate. The detailed results below retain their scales, configurations and source dates.

Evaluator	Test series	Composite indices	Recorded active measurements
Artificial Analysis	10	1	13
Vals AI	27	2	211
LMArena	4	0	43
LiveBench	0	1	8
Epoch AI	6	0	12
Scale AI Evals	3	0	3

Coverage is not a quality ranking. Overlapping tests and multiple settings do not count as independent experiments. Vendor claims, historical and disputed results are excluded from this overview and remain labelled in the detailed evidence.

Published benchmarks

Reasoning

LiveBench

A regularly refreshed mix of objectively scored tasks across seven categories. The composite applies only to the specific release and its weighting.

Sub-task	Claude Fable 5 (This model)	GPT-5.6 Sol	GPT-5.6 Terra	GPT-5.5
Agentic coding	62.2 (best value in this row)	56.2	54.9	54
Overall	82.97 (best value in this row)	81.05	77.94	80.19
Coding	86 (best value in this row)	83.9	78.2	82.1
Data analysis	80.5	79.8	79.3	81.6 (best value in this row)
Instruction following	75.8 (best value in this row)	71.8	64.6	70.7
Language	90.7 (best value in this row)	87.7	82.9	87.4
Mathematics	96	96.2 (best value in this row)	94.9	95.9
Reasoning	89.7	91.7 (best value in this row)	90.6	89.7

Unit: pointshigher is betterAs of September 2, 2026Source: LiveBenchbars scaled within each row, not across rows

EnigmaEval

Overall

1,161 complex, often multimodal puzzles test creative problem solving beyond well-defined academic tasks. Puzzle solving is a narrow specialty and does not replace professional or everyday task evaluation.

1.Claude Fable 539.28%

2.GPT-5.6 Sol37.12%

3.Gemini 3.1 Pro Preview36.78%

4.Gemini 3.5 Flash25.41%

5.Claude Opus 4.823.51%

higher is betterAs of August 26, 2026Scale: 0% - 100%

Source: Scale AI

|

gradually.ai

Methodology and context

Comparison cohort: scale-enigma-eval-2026-07-23
Metric: pass@1 accuracy
Scale: 0-100
Participants: 5
Methodology and context: Complex multimodal puzzles with exact answer matching and one attempt.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-high. Evaluator: Scale AI. Dataset type: public. Leaderboard date: 2026-07-23. Attempts: pass@1. Reasoning effort: high. Score display: 39.28% ± 2.8. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-de06ddcfa8a8. Source payload hash: de06ddcfa8a81a2242bdcf8cae4bc1d2e0d1d9e896cb450a19ce0227b96075a7. Parser version: curated-v1. Imported at: 2026-10-15

Source: Scale AI

PRBench Finance

Overall

600 realistic finance questions with weighted criteria, including a 300-question hard subset. An o4-mini judge scores the responses, so domain correctness and operational reliability remain separate questions.

1.Claude Fable 553.86%

2.Claude Opus 4.653.28%

3.GPT-551.32%

4.GPT-5 Pro51.06%

5.GPT-5.6 Sol50.45%

6.o3-pro49.08%

7.o347.69%

8.Claude Opus 4.546.16%

9.GPT-5.445.63%

10.Claude Sonnet 4.543.79%

11.Gemini 3.1 Pro Preview41.87%

12.o4-mini39.22%

13.Gemini 2.5 Pro38.92%

14.Gemini 2.5 Flash38.41%

15.GPT-4.134.32%

16.Llama 4 Maverick22.36%

higher is betterAs of August 26, 2026Scale: 0% - 100%

Source: Scale AI

|

gradually.ai

Methodology and context

Comparison cohort: scale-prbench-finance-full
Metric: weighted rubric score
Scale: 0-100
Participants: 16
Methodology and context: 600 finance tasks with task-specific rubrics and o4-mini as judge.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-max. Evaluator: Scale AI. Dataset type: public. Attempts: pass@1. Reasoning effort: max. Score display: 53.86% ± 0.14. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-bf274b71f97c. Source payload hash: bf274b71f97c40f0911be8c49cc3640b9020b7e8f195d1da9823e3c03787cbe3. Parser version: curated-v1. Imported at: 2026-10-15

Source: Scale AI

PRBench Legal

Overall

500 realistic legal questions with weighted criteria across jurisdictions and practice areas. The benchmark measures professional response quality but cannot replace legal review or human accountability.

1.Claude Fable 552.56%

2.Claude Opus 4.652.27%

3.GPT-5.6 Sol50.5%

4.GPT-5 Pro49.89%

5.o3-pro49.67%

6.GPT-548.96%

7.o348.57%

8.GPT-5.444.35%

9.Claude Opus 4.544.21%

10.Gemini 3.1 Pro Preview44.02%

11.Gemini 2.5 Pro41.43%

12.Gemini 2.5 Flash41.02%

13.Claude Sonnet 4.540.84%

14.o4-mini38.11%

15.GPT-4.136.48%

16.Llama 4 Maverick24.84%

higher is betterAs of August 26, 2026Scale: 0% - 100%

Source: Scale AI

|

gradually.ai

Methodology and context

Comparison cohort: scale-prbench-legal-full
Metric: weighted rubric score
Scale: 0-100
Participants: 16
Methodology and context: 500 legal tasks with task-specific rubrics and o4-mini as judge.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-max. Evaluator: Scale AI. Dataset type: public. Attempts: pass@1. Reasoning effort: max. Score display: 52.56% ± 0.54. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-6eb497f65483. Source payload hash: 6eb497f65483f7d7f740b0e8e20ca95ea1a57c2bd1a03575b3448bb5e468d88f. Parser version: curated-v1. Imported at: 2026-10-15

Source: Scale AI

SWE-Marathon v1.0

Binary resolution, pass@1

20 multi-hour repository tasks test very long software work with binary end-to-end resolution. Scores belong to the v1.0 agent setup. SWE-Marathon v1.1 has been released since July 2026. This dataset records v1.0 results only.

Note (Claude Fable 5): Two refusals and three unavailable provider-ban trials were counted as zero.

2.Claude Opus 4.826%

3.Claude Fable 524%

4.Claude Opus 4.716%

6.Gemini 3.5 Flash7%

higher is betterAs of August 9, 2026Scale: 0% - 100%

Source: SWE-Marathon

|

gradually.ai

Methodology and context

Comparison cohort: swe-marathon-v1-0:official:pass1
Metric: binary resolution pass@1
Scale: 0-100
Participants: 6
Methodology and context: 20 multi-hour tasks. The named coding agent is part of every result row.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: SWE-Marathon. Dataset type: public. Leaderboard date: 2026-06-05. Agent: Claude Code. Source ID: swe-marathon. Evidence capture ID: curated-swe-marathon-2026-08-09-0fdff812abab. Source payload hash: 0fdff812abab3328557b063f7f54c9456293239f1de9db283ce69b46a21b63f1. Parser version: curated-v1. Imported at: 2026-10-15

Source: SWE-Marathon

Recorded measurement details

LiveBench: Agentic coding (Show measurement, test conditions, and source)

Source value: 62.2
Score: 62.2
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: agentic-coding
Task label: Agentic coding
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-agentic-coding-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Agentic coding category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-94a9090eb4a5. Source payload hash: 94a9090eb4a5441e58ab80974bafcbdfa285cf61fccc1c6d3fe77758c4e576bd. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Agentic coding category inside the same LiveBench release.

Source LiveBench

LiveBench: Overall (Show measurement, test conditions, and source)

Source value: 82.97
Score: 82.97
Metric: composite score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: 23 objectively scored tasks across seven categories.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-122dcbf95216. Source payload hash: 122dcbf95216ffd69bb359d05ff57d48070d378865a810b62a69f54ca644afed. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: 23 objectively scored tasks across seven categories.

Source LiveBench

LiveBench: Coding (Show measurement, test conditions, and source)

Source value: 86
Score: 86
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: coding
Task label: Coding
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-coding-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Coding category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-512c15649aa6. Source payload hash: 512c15649aa6363ed1ebb0b149ba6f57b9c66675d1aa3e58262736070fea8b32. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Coding category inside the same LiveBench release.

Source LiveBench

LiveBench: Data analysis (Show measurement, test conditions, and source)

Source value: 80.5
Score: 80.5
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: data-analysis
Task label: Data analysis
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-data-analysis-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Data analysis category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-fd85d8652330. Source payload hash: fd85d86523309d4bc8c34b7f1becc0755dad6f072b5c46c4fd8ad17e89a01101. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Data analysis category inside the same LiveBench release.

Source LiveBench

LiveBench: Instruction following (Show measurement, test conditions, and source)

Source value: 75.8
Score: 75.8
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: instruction-following
Task label: Instruction following
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-instruction-following-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Instruction following category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-c906e5c563af. Source payload hash: c906e5c563af6f19d4f25800569607338491c293f72c448edd09daffd68bec65. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Instruction following category inside the same LiveBench release.

Source LiveBench

LiveBench: Language (Show measurement, test conditions, and source)

Source value: 90.7
Score: 90.7
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: language
Task label: Language
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-language-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Language category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-eb8cc58ecaaa. Source payload hash: eb8cc58ecaaa71861acc480b00537a09b06bb9ee310b1b24a9a8754cc086796d. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Language category inside the same LiveBench release.

Source LiveBench

LiveBench: Mathematics (Show measurement, test conditions, and source)

Source value: 96
Score: 96
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: mathematics
Task label: Mathematics
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-mathematics-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Mathematics category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-df0ccd787b7a. Source payload hash: df0ccd787b7ac45851c25541373f20892850c2ed4071fd6a44700630ad23e50c. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Mathematics category inside the same LiveBench release.

Source LiveBench

LiveBench: Reasoning (Show measurement, test conditions, and source)

Source value: 89.7
Score: 89.7
Metric: category score
Unit: points
Benchmark version: 2026-06-25
Category: reasoning
Task: reasoning
Task label: Reasoning
Direction: Higher is better
Comparison cohort: livebench-2026-06-25-reasoning-max
Source type: official-leaderboard
Evaluator: LiveBench
Status: active
Retrieved at: 2026-09-02
Methodology: Reasoning category inside the same LiveBench release.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: LiveBench. Dataset type: public. Leaderboard date: 2026-06-25. Source ID: livebench. Evidence capture ID: curated-livebench-2026-09-02-0a36eddee978. Source payload hash: 0a36eddee978a774e485d6988e5ef8add79e2adf8105068cee59eb1e427301b0. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Reasoning category inside the same LiveBench release.

Source LiveBench

EnigmaEval: Overall (Show measurement, test conditions, and source)

Source value: 39.28
Score: 39.28
Metric: pass@1 accuracy
Unit: %
Benchmark version: open-source release 2026-07-23
Category: reasoning
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: scale-enigma-eval-2026-07-23
Confidence interval: 2.8
Source type: official-leaderboard
Evaluator: Scale AI
Status: active
Retrieved at: 2026-08-26
Methodology: Complex multimodal puzzles with exact answer matching and one attempt.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-high. Evaluator: Scale AI. Dataset type: public. Leaderboard date: 2026-07-23. Attempts: pass@1. Reasoning effort: high. Score display: 39.28% ± 2.8. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-de06ddcfa8a8. Source payload hash: de06ddcfa8a81a2242bdcf8cae4bc1d2e0d1d9e896cb450a19ce0227b96075a7. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: Complex multimodal puzzles with exact answer matching and one attempt.

Source Scale AI

PRBench Finance: Overall (Show measurement, test conditions, and source)

Source value: 53.86
Score: 53.86
Metric: weighted rubric score
Unit: %
Benchmark version: full finance set
Category: finance
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: scale-prbench-finance-full
Confidence interval: 0.14
Source type: official-leaderboard
Evaluator: Scale AI
Status: active
Retrieved at: 2026-08-26
Methodology: 600 finance tasks with task-specific rubrics and o4-mini as judge.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-max. Evaluator: Scale AI. Dataset type: public. Attempts: pass@1. Reasoning effort: max. Score display: 53.86% ± 0.14. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-bf274b71f97c. Source payload hash: bf274b71f97c40f0911be8c49cc3640b9020b7e8f195d1da9823e3c03787cbe3. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: 600 finance tasks with task-specific rubrics and o4-mini as judge.

Source Scale AI

PRBench Legal: Overall (Show measurement, test conditions, and source)

Source value: 52.56
Score: 52.56
Metric: weighted rubric score
Unit: %
Benchmark version: full legal set
Category: legal
Task: overall
Task label: Overall
Direction: Higher is better
Comparison cohort: scale-prbench-legal-full
Confidence interval: 0.54
Source type: official-leaderboard
Evaluator: Scale AI
Status: active
Retrieved at: 2026-08-26
Methodology: 500 legal tasks with task-specific rubrics and o4-mini as judge.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Model snapshot: claude-fable-5-max. Evaluator: Scale AI. Dataset type: public. Attempts: pass@1. Reasoning effort: max. Score display: 52.56% ± 0.54. Source ID: scale-ai. Evidence capture ID: curated-scale-ai-2026-08-26-6eb497f65483. Source payload hash: 6eb497f65483f7d7f740b0e8e20ca95ea1a57c2bd1a03575b3448bb5e468d88f. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: 500 legal tasks with task-specific rubrics and o4-mini as judge.

Source Scale AI

SWE-Marathon v1.0: Binary resolution, pass@1 (Show measurement, test conditions, and source)

Source value: 24
Score: 24
Metric: binary resolution pass@1
Unit: %
Benchmark version: v1.0 leaderboard
Category: coding
Task: pass1
Task label: Binary resolution, pass@1
Direction: Higher is better
Comparison cohort: swe-marathon-v1-0:official:pass1
Source type: official-leaderboard
Evaluator: SWE-Marathon
Status: active
Retrieved at: 2026-08-09
Methodology: 20 multi-hour tasks. The named coding agent is part of every result row.. Source model ID: curated/claude-fable-5. Source model name: Claude Fable 5. Provider reported by source: Anthropic. Evaluator: SWE-Marathon. Dataset type: public. Leaderboard date: 2026-06-05. Agent: Claude Code. Source ID: swe-marathon. Evidence capture ID: curated-swe-marathon-2026-08-09-0fdff812abab. Source payload hash: 0fdff812abab3328557b063f7f54c9456293239f1de9db283ce69b46a21b63f1. Parser version: curated-v1. Imported at: 2026-10-15
Comparison methodology: 20 multi-hour tasks. The named coding agent is part of every result row.
Context: Two refusals and three unavailable provider-ban trials were counted as zero.

Source SWE-Marathon

Page 1 of 25

Predecessor and successor

Earlier model generation: Claude Mythos Preview (Official documentation)
Official documentation
Model lineage evidence
The same guide covers migrating from Claude Mythos Preview to either target and states "Claude Fable 5 offers the same capabilities [as Claude Mythos 5] and does not require access approval"; Anthropic's explicit word "successor" is used only for the Mythos 5 side of this pairing, so Fable 5's link to Mythos Preview is recorded as nearest-profile rather than direct.
Anthropic "Migrating to Claude Mythos 5 and Claude Fable 5" guide (2026-09-19)
Successor: Claude Fable 5.1
Model lineage evidence
The guide has a dedicated section "Migrating to Claude Fable 5.1 from Claude Fable 5" stating the API surface, limits, pricing, tokenizer and thinking behavior "all match Claude Fable 5.".
Anthropic "Migrating to Claude Fable 5.1 and Claude Mythos 5.1" guide (2026-09-19)

Sources and data date

Type	Evidence and data date
Speed	Anthropic, Adaptive reasoning, max effort, Opus 4.8 fallback. Output speed after the first streaming chunk, not total response time. Retrieved September 7, 2026. Source: Artificial Analysis · Speed (Artificial Analysis)Speed measurement methodology
Price	Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. Source: Pricing · API pricing
Input	Documented modalities. Retrieved September 7, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Fable 5 model overview · Editorial description reviewed October 2, 2026
Output	Documented modalities. Retrieved September 7, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Fable 5 model overview · Editorial description reviewed October 2, 2026
Research status	Research date September 8, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not found in the inspected sources	parameters-billions Original wording Anthropic does not publish parameter counts for this model.
Not found in the inspected sources	active-parameters-billions Original wording Anthropic does not publish active parameter counts for this model.
Unresolved	profile-benchmarks-identity Original wording The public table combines the same underlying Mythos 5/Fable 5 model and reports the higher score where safeguards differ; it does not expose separate exact per-variant cells.
Unresolved	profile-benchmarks Original wording Official launch table combines Mythos 5 / Fable 5 and reports the higher variant score; retained as separate context, not an exact per-profile benchmark.
Additional source	Model metadata · Anthropic, Claude Fable 5 and Claude Mythos 5 · Editorial description reviewed October 2, 2026
Additional source	Context window data
Additional source	Knowledge cutoff data
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	Artificial Analysis (retrieved October 19, 2026)
Additional source	LiveBench (retrieved September 2, 2026)
Additional source	OpenAI (retrieved September 2, 2026)
Additional source	Scale AI (retrieved August 26, 2026)
Additional source	Scale AI (retrieved August 26, 2026)
Additional source	Scale AI (retrieved August 26, 2026)
Additional source	SWE-Marathon (retrieved August 9, 2026)
Additional source	Terminal-Bench (retrieved October 15, 2026)
Additional source	LMArena (retrieved October 15, 2026)
Additional source	LMArena (retrieved October 15, 2026)
Additional source	LMArena (retrieved October 15, 2026)
Additional source	LMArena (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved August 11, 2026; October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Vals AI (retrieved October 15, 2026)
Additional source	Epoch AI benchmark run dataset snapshot (retrieved October 19, 2026)
Additional source	Anthropic, Redeploying Fable 5 · Editorial description reviewed October 2, 2026
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Ollama: cumulative downloads for the popularity index
Additional source	Replicate: cumulative runs for the popularity index
Additional source	Google Trending Now: limited feed selection for Germany and the US
Additional source	Creative Commons Attribution 4.0
Additional source	Anthropic "Migrating to Claude Mythos 5 and Claude Fable 5" guide (2026-09-19)
Additional source	Anthropic "Migrating to Claude Fable 5.1 and Claude Mythos 5.1" guide (2026-09-19)