modelbench日本語

Claude Fable 5.2

AnthropicCurrent

Input price
$8.00
Per 1 million tokens (USD)
Output price
$40.00
Per 1 million tokens (USD)
Effort levels
5total
Default: High
Benchmark results
51total
Benchmark types: 26

Model information

API model ID
claude-fable-5-2
release date
2026-11-09
effort
Low / Med / High (default) / Xhigh / Max
Effort source
Based on Anthropic effort documentation (platform.claude.com/docs/en/build-with-claude/effort).
Claude Code
available
Codex
unavailable
Availability notes
`/model fable` (resolves to Fable 5.2 as of v2.1.321). The Claude Code default effort is medium (API default is high). Subscriptions may be billed in usage credits. Codex's official model list (learn.chatgpt.com/docs/models) contains only OpenAI models, so codex is false (connections via a custom provider are not counted as official options).
Availability source
https://code.claude.com/docs/en/model-config

Price (USD per 1 million tokens)

Prices do not vary by effort; token consumption does.

input$8.00
output$40.00
cache read$0.20
cache write$10.00

cacheWritePerMTok is for a 5-minute cache write. 1h write $16. cache read is 0.025x the base input. fast mode $16/$80. 1M context is offered at standard pricing.

Claude Fable 5.2 | $8 / MTok | $10 / MTok | $16 / MTok | $0.20 / MTok | $40 / MTok

Source: https://platform.claude.com/docs/en/about-claude/pricing

Benchmark results

BenchmarkeffortScoreSource cost (metric)measured/reported bysourceConditions and extraction method
Terminal-Bench 4.0MaxMax67.6%—AnthropicClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Claude Code --bare, max, 66 tasks × 10 trials. safeguards enabled, default fallback (2.5% of requests, 10% of trials). No internet; required resources cached in advance. Standard error ±1.8 pt. Cost not stated.transcribed from text
Source excerpt67.6%Location §8.5 “Terminal-Bench 4.0”, body text, Fable 5.2 score
Terminal-Bench-Science 0.1not statednot stated74.6%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt74.6%Location Table, Terminal-Bench-Science 0.1 row: 74.6
FrontierCode 2.0LowLow34.2%$2.10Cost per taskAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptFrontierCode 2.0 — Accuracy vs. cost: "Fable 5.2 · Low: 34.2% $2.10"
FrontierCode 2.0MedMed37.6%$2.48Cost per taskAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptFrontierCode 2.0 — Accuracy vs. cost: "Fable 5.2 · Med: 37.6% $2.48"
FrontierCode 2.0High (default)High45.3%$6.72Cost per taskAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptFrontierCode 2.0 — Accuracy vs. cost: "Fable 5.2 · High: 45.3% $6.72"
FrontierCode 2.0XhighXhigh49.1%$14.19Cost per taskAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptFrontierCode 2.0 — Accuracy vs. cost: "Fable 5.2 · Xhigh: 49.1% $14.19"
FrontierCode 2.0MaxMax45.6%$44.70Cost per taskAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptFrontierCode 2.0 — Accuracy vs. cost: "Fable 5.2 · Max: 45.6% $44.70"
FrontierSWE 2not statednot stated69.9%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt69.9%Location Table, FrontierSWE v2 row: 69.9
CursorBench 4.0LowLow53.2%—CursorClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Independently run and reported by Cursor in its production agent harness. A single result per effort; no uncertainty interval. Additional prompt and harness improvements prevent direct comparison with values in the earlier Opus 5.5 and Sonnet 5.5 system cards. Numeric cost not stated in the body text.transcribed from text
Source excerpt53.2%Location §8.8 “CursorBench 4.0”, body text, Fable 5.2 scores by effort
CursorBench 4.0MedMed54.8%—CursorClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Independently run and reported by Cursor in its production agent harness. A single result per effort; no uncertainty interval. Additional prompt and harness improvements prevent direct comparison with values in the earlier Opus 5.5 and Sonnet 5.5 system cards. Numeric cost not stated in the body text.transcribed from text
Source excerpt54.8%Location §8.8 “CursorBench 4.0”, body text, Fable 5.2 scores by effort
CursorBench 4.0High (default)High56.9%—CursorClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Independently run and reported by Cursor in its production agent harness. A single result per effort; no uncertainty interval. Additional prompt and harness improvements prevent direct comparison with values in the earlier Opus 5.5 and Sonnet 5.5 system cards. Numeric cost not stated in the body text.transcribed from text
Source excerpt56.9%Location §8.8 “CursorBench 4.0”, body text, Fable 5.2 scores by effort
CursorBench 4.0XhighXhigh56.3%—CursorClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Independently run and reported by Cursor in its production agent harness. A single result per effort; no uncertainty interval. Additional prompt and harness improvements prevent direct comparison with values in the earlier Opus 5.5 and Sonnet 5.5 system cards. Numeric cost not stated in the body text.transcribed from text
Source excerpt56.3%Location §8.8 “CursorBench 4.0”, body text, Fable 5.2 scores by effort
CursorBench 4.0MaxMax60.0%—CursorClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Independently run and reported by Cursor in its production agent harness. A single result per effort; no uncertainty interval. Additional prompt and harness improvements prevent direct comparison with values in the earlier Opus 5.5 and Sonnet 5.5 system cards. Numeric cost not stated in the body text.transcribed from text
Source excerpt60.0%Location §8.8 “CursorBench 4.0”, body text, Fable 5.2 scores by effort
GDPval-AA v2.1XhighXhigh1,751—Artificial AnalysisClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Anthropic publishes values independently measured by Artificial Analysis. GDPval-AA v2.1, 220 tasks, Elo from blind pairwise comparisons. Cost not stated.transcribed from text
Source excerpt1751Location §8.14.3 “GDPval-AA v2.1”, body text, Fable 5.2 Elo
GDPval-AA v2.1MaxMax1,831—Artificial AnalysisClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Anthropic publishes values independently measured by Artificial Analysis. GDPval-AA v2.1, 220 tasks, Elo from blind pairwise comparisons. Cost not stated.transcribed from text
Source excerpt1831Location §8.14.3 “GDPval-AA v2.1”, body text, Fable 5.2 Elo
AA-Briefcase v1.1(Elo)not statednot stated1,785—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt1785Location Table, AA-Briefcase v1.1 (Elo) row: 1785
AA-Briefcase v1.1(rubric)not statednot stated66.1%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt66.1%Location Table, AA-Briefcase v1.1 (rubric) row: 66.1
AutomationBenchnot statednot stated51.4%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt51.4%Location Table, AutomationBench row: 51.4
Humanity's Last Exam (with tools)LowLow68.4%$0.19Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptHumanity’s Last Exam (with tools) — Accuracy vs. cost: "Fable 5.2 · Low: 68.4% $0.19"
Humanity's Last Exam (with tools)MedMed68.6%$0.21Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptHumanity’s Last Exam (with tools) — Accuracy vs. cost: "Fable 5.2 · Med: 68.6% $0.21"
Humanity's Last Exam (with tools)High (default)High71.5%$0.63Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptHumanity’s Last Exam (with tools) — Accuracy vs. cost: "Fable 5.2 · High: 71.5% $0.63"
Humanity's Last Exam (with tools)XhighXhigh73.4%$1.48Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptHumanity’s Last Exam (with tools) — Accuracy vs. cost: "Fable 5.2 · Xhigh: 73.4% $1.48"
Humanity's Last Exam (with tools)MaxMax74.2%$4.51Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptHumanity’s Last Exam (with tools) — Accuracy vs. cost: "Fable 5.2 · Max: 74.2% $4.51"
Humanity's Last Exam (without tools)MaxMax73.2%—AnthropicClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Default settings for Table 8.1.A: adaptive thinking, max effort, default sampling, averaged over 5 trials. without tools. Cost not stated.transcribed from table
Source excerpt73.2Location Table 8.1.A “Evaluation summary”, Humanity’s Last Exam / No tools row, own model column
OSWorld 2.1 (offline set, partial)not statednot stated89.3%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt89.3%Location Table, OSWorld 2.1 (offline set, partial) row: 89.3
Chartography (with tools)MaxMax91.6%—AnthropicClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Run internally by Anthropic. adaptive thinking, max, with tools, averaged over 5 runs. The grader is Gemini 3.5 Flash. Cost not stated.transcribed from text
Source excerpt91.6%Location §8.13.1 “Chartography”, body text, Fable 5.2 with tools score
Chartography (without tools)LowLow64.2%$0.03Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptChartography (no tools) — Accuracy vs. cost: "Fable 5.2 · Low: 64.2% $0.03"
Chartography (without tools)MedMed66.4%$0.03Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptChartography (no tools) — Accuracy vs. cost: "Fable 5.2 · Med: 66.4% $0.03"
Chartography (without tools)High (default)High69.8%$0.07Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptChartography (no tools) — Accuracy vs. cost: "Fable 5.2 · High: 69.8% $0.07"
Chartography (without tools)XhighXhigh71.4%$0.15Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptChartography (no tools) — Accuracy vs. cost: "Fable 5.2 · Xhigh: 71.4% $0.15"
Chartography (without tools)MaxMax73.4%$0.75Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptChartography (no tools) — Accuracy vs. cost: "Fable 5.2 · Max: 73.4% $0.75"
WANDR(relaxed score)LowLow71.4%$12.66Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text
Source excerptWANDR (relaxed score) — Accuracy vs. cost: "Fable 5.2 · Low: 71.4% $12.66"
WANDR(relaxed score)MedMed73.6%$15.71Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text
Source excerptWANDR (relaxed score) — Accuracy vs. cost: "Fable 5.2 · Med: 73.6% $15.71"
WANDR(relaxed score)High (default)High79.3%$29.20Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text
Source excerptWANDR (relaxed score) — Accuracy vs. cost: "Fable 5.2 · High: 79.3% $29.20"
WANDR(relaxed score)XhighXhigh80.6%$39.84Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text
Source excerptWANDR (relaxed score) — Accuracy vs. cost: "Fable 5.2 · Xhigh: 80.6% $39.84"
WANDR(relaxed score)MaxMax82.1%$53.11Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text
Source excerptWANDR (relaxed score) — Accuracy vs. cost: "Fable 5.2 · Max: 82.1% $53.11"
BenchCAD 2.1 (with tools)LowLow26.9$1.89Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptBenchCAD 2.1 (with tools) — Accuracy vs. cost: "Fable 5.2 · Low: 26.9 $1.89"
BenchCAD 2.1 (with tools)MedMed32.4$2.66Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptBenchCAD 2.1 (with tools) — Accuracy vs. cost: "Fable 5.2 · Med: 32.4 $2.66"
BenchCAD 2.1 (with tools)High (default)High61$9.99Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptBenchCAD 2.1 (with tools) — Accuracy vs. cost: "Fable 5.2 · High: 61.0 $9.99"
BenchCAD 2.1 (with tools)XhighXhigh74.1$23.23Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptBenchCAD 2.1 (with tools) — Accuracy vs. cost: "Fable 5.2 · Xhigh: 74.1 $23.23"
BenchCAD 2.1 (with tools)MaxMax83.5$50.77Cost per attemptAnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text
Source excerptBenchCAD 2.1 (with tools) — Accuracy vs. cost: "Fable 5.2 · Max: 83.5 $50.77"
RoboDojo Simnot statednot stated64.9%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted. ‡ highest-available effort, run internally.transcribed from table
Source excerpt64.9%Location Table, RoboDojo Sim row: 64.9
Remote Labor Index(RLI)High (default)High41.3%—AnthropicIntroducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table
Source excerpt41.3%Location Table, Remote Labor Index (RLI) row: 41.3 high
HealthBench Professional (length-adjusted score)MaxMax67.7—Artificial AnalysisClaude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14Vendor-reported value. Independently run by Artificial Analysis at Anthropic’s request. max, length-adjusted score, GPT-5.4 low grader. fallback enabled. Not comparable with internal runs in earlier Claude system cards (Opus 4.8 grader). Cost not stated.transcribed from text
Source excerpt67.7%Location §8.16.2 “HealthBench Professional (Artificial Analysis)”, body text, the model’s length-adjusted score in parentheses
GPQA DiamondMaxMax93.9%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Fable 5.2 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptgpqa_diamond.csv: Model version=claude-fable-5-2_max, Best score (across scorers)=0.9387626262626263, stderr=0.016253431037635206, Started at=2026-11-06T01:14:45.000Z
FrontierMath Tiers 1-3(v2)MaxMax95.1%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Fable 5.2 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptfrontiermath_tiers_1_3_v2.csv: Model version=claude-fable-5-2_max, Best score (across scorers)=0.9508771929824561, stderr=0.012824623769055387, Started at=2026-11-06T01:14:45.000Z
FrontierMath Tier 4(v2)MaxMax100.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Fable 5.2 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptfrontiermath_tier_4_v2.csv: Model version=claude-fable-5-2_max, Best score (across scorers)=1.0, stderr=0.0, Started at=2026-11-06T01:14:45.000Z
OTIS Mock AIME 2024-2025MaxMax99.7%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Fable 5.2 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptotis_mock_aime_2024_2025.csv: Model version=claude-fable-5-2_max, Best score (across scorers)=0.9972222222222222, stderr=0.002777777777777778, Started at=2026-11-06T01:14:45.000Z
SimpleQA VerifiedMaxMax94.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Fable 5.2 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptsimpleqa_verified.csv: Model version=claude-fable-5-2_max, Best score (across scorers)=0.94, stderr=0.007513751157474972, Started at=2026-11-06T01:14:45.000Z
Chess Puzzles(Epoch AI)MaxMax79.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Fable 5.2 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table
Source excerptchess_puzzles.csv: Model version=claude-fable-5-2_max, Best score (across scorers)=0.79, stderr=0.040936018074033236, Started at=2026-11-06T01:14:45.000Z
Epoch Capabilities Index(ECI)not statednot stated173.44—Epoch AIEpoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)third-party · Retrieved 2026-11-12A per-model metric that combines results across multiple benchmarks and multiple efforts (not per effort). Calculated by Epoch AI.transcribed from table
Source excerpteci_scores.csv: Model=Claude Fable 5.2, eci=173.44, eci_ci_low=169.93, eci_ci_high=177.49

Related rankings

References