Claude Fable 5.2在Artificial Analysis获65分创最高

Artificial Analysis

Anthropic

Claude Fable 5.2

Proprietary model

Released November 2026

Claude Fable 5.2 (Max, Default Fallback) Intelligence, Performance & Price Analysis

Try it out API Provider Benchmarks

Model summary

IntelligenceUpdated

Artificial Analysis Intelligence Index

4 out of 4 units for Intelligence.

Performance

Time to first answer token 219.85sOutput speed 98 tok/s

Time per Intelligence Index task (minutes)

4 out of 4 units for Performance.

In $8.00Out $40.00Cache discount 98%

$26.54

Cost per Intelligence Index task

4 out of 4 units for Cost.

Verbosity

Input tokens 28.3BOutput/input 0.97%

Output tokens from Intelligence Index

4 out of 4 units for Verbosity.

&]:visible [[data-state=open]>&]:grid-rows-[1fr] [[data-state=open]>&]:opacity-100">

Claude Fable 5.2 (Max, Default Fallback) is amongst the leading models in intelligence, but particularly expensive when comparing to other models of similar price. It's also notably slow per task and very verbose. The model supports text and image input, outputs text, and has a 1M tokens context window.

Claude Fable 5.2 (Max, Default Fallback) scores 65 on the Artificial Analysis Intelligence Index, placing it well above average among comparable models (median: 40). When evaluating the Intelligence Index, it generated 275M tokens, which is very verbose in comparison to the median of 72.8M.

Pricing for Claude Fable 5.2 (Max, Default Fallback) is $8.00 per 1M input tokens (expensive, median: $2.00) and $40.00 per 1M output tokens (expensive, median: $10.00). On average, it costs $26.54 per task to evaluate Claude Fable 5.2 (Max, Default Fallback) on the Intelligence Index.

At 46.4 minutes per Intelligence Index task, Claude Fable 5.2 (Max, Default Fallback) is notably slow compared with the median of 12.6 minutes for comparable models (lower is better).

&]:visible [[data-state=open]>&]:grid-rows-[1fr] [[data-state=open]>&]:opacity-100">

ReasoningYes

This page shows the reasoning version of this model.

A non-reasoning variant may also exist.

Input modality

Supports: text and image

Output modality

Supports: text

Context window1M

~1500 A4 pages of size 12 Arial font

&]:visible [[data-state=open]>&]:grid-rows-[1fr] [[data-state=open]>&]:opacity-100">

Metrics are compared against models of the same class:

· Non-reasoning models → compared only with other non-reasoning models

· Reasoning models → compared across both reasoning and non-reasoning

· Open weights models → compared only with other open weights models of the same size class:

· Tiny: ≤4B parameters

· Small: 4B–40B parameters

· Medium: 40B–150B parameters

· Large: >150B parameters

· Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio:

· <$0.15 per 1M tokens

· $0.15–$1 per 1M tokens

· >$1 per 1M tokens

Updated

Intelligence

Artificial Analysis Intelligence Index · Higher is better

No publicly available API

Time per Task

Inference time (minutes) per Intelligence Index task · Lower is better

No publicly available API

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better

No publicly available API

IntelligenceUpdated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

No publicly available API

&]:visible [[data-state=open]>&]:grid-rows-[1fr] [[data-state=open]>&]:opacity-100">

Artificial Analysis Intelligence Index v5.0.1 includes: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

No publicly available API

&]:visible [[data-state=open]>&]:grid-rows-[1fr] [[data-state=open]>&]:opacity-100">

Artificial Analysis Intelligence Index v5.0.1 includes: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

&]:visible [[data-state=open]>&]:grid-rows-[1fr] [[data-state=open]>&]:opacity-100">

Indicates whether the model weights are available. Models are labelled as 'Commercial Use Restricted' if commercial use is limited by conditions, and as 'Non-commercial' if the license prohibits commercial use.

Intelligence Index vs. Cost per Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task