ZMIME
Anthropic · Released Nov 9, 2026

Claude Fable 5.2

Claude model for demanding reasoning and long-horizon agentic work.

  • New
  • Proprietary
  • Reasoning
  • Tool calling
  • Vision
Capability
173.5ECI · #2 of 158
Input
$8.00per 1M tokens
Output
$40.00per 1M tokens
Context
1M128K max output
Pricing

What it costs

Input
$8.00per million tokens
Output
$40.00per million tokens
Cached input
$0.20per million tokens
Cache write
$10.00per million tokens
Blended (3:1)
$16.00pricier than 96% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$36.00
Team assistant25M input + 5M output tokens$400.00
Production app250M input + 50M output tokens$4,000

Official Anthropic API price, as listed on models.dev.

Capability

Independent benchmarks

173.5Epoch Capabilities Index
#2 of 158 scored models
68Median 147174

The shaded band is Epoch AI’s confidence range (170.1–177.4); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions · max setting96.0%
  • FrontierMath Tiers 1–3Research-level mathematics · max setting95.1%
  • OTIS Mock AIME 2024–2025Competition mathematics · max setting99.7%
  • SimpleQA VerifiedShort factual questions · max setting94.0%

Scores from Epoch AI, run independently of Anthropic.

Specifications

The details

Lab
Anthropic
Released
Nov 9, 2026
Knowledge cutoff
Jun 2026
Context window
1,000,000 tokens
Max output
128,000 tokens
Inputs
Text, Images, PDFs
Output
Text
Reasoning
Adjustable effort low · medium · high · xhigh · max
Tool calling
Yes
Structured output
Yes
Weights
Proprietary
API model ID
claude-fable-5-2on Anthropic
Availability
25 API providerslisted on models.dev

Anthropic model documentation

Lab-reported

What Anthropic claims

Published by the lab at launch. Settings vary, so compare these only with care.

BenchmarkScoreSettingSource
FrontierCode v2.049.1xhigh effort; safeguards onSource
FrontierSWE v269.9highest available effort; safeguards onSource
AutomationBench51.4highest available effort; safeguards onSource
Terminal-Bench-Science v0.174.6highest available effort; safeguards onSource
WANDR82.1highest available effort; safeguards onSource
OSWorld v2.189.3highest available effort; safeguards onSource
RoboDojo Sim64.9 average success ratemax effort; internal runSource
AA-Briefcase v1.166.1highest available effort; safeguards onSource
AA-Briefcase v1.11785 Elohighest available effort; safeguards onSource
Remote Labor Index41.3high effort; safeguards onSource
Humanity's Last Exam74.2with tools; highest available effort; safeguards onSource
Chartography73.4no tools; highest available effort; safeguards onSource
Same lab

More from Anthropic

Questions

About Claude Fable 5.2

How much does Claude Fable 5.2 cost?

Claude Fable 5.2 costs $8.00 input / $40.00 output per million tokens (official Anthropic API price). Cached input is $0.20 per million tokens, and writing to the cache costs $10.00. At a 3:1 input-to-output mix that is $16.00 per million tokens, more expensive than 96% of the 469 priced models we track.

What is the context window of Claude Fable 5.2?

Claude Fable 5.2 accepts up to 1,000,000 tokens per request and can write up to 128,000 tokens in one response.

How good is Claude Fable 5.2?

Epoch AI gives Claude Fable 5.2 a Capabilities Index score of 173.5 (likely range 170.1–177.4), ranking it #2 of 158 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 96.0%, FrontierMath Tiers 1–3 95.1%, OTIS Mock AIME 2024–2025 99.7%, SimpleQA Verified 94.0%.

Is Claude Fable 5.2 open source?

No. Claude Fable 5.2 is proprietary; you use it through Anthropic’s API or partner platforms.

What inputs does Claude Fable 5.2 support?

Claude Fable 5.2 accepts text, images and PDFs and replies in text. It is a reasoning model with low, medium, high, xhigh and max effort settings, supports tool calling and can return structured JSON output.

When was Claude Fable 5.2 released?

Anthropic released Claude Fable 5.2 on Nov 9, 2026. Its training data runs to Jun 2026.

What are the best alternatives to Claude Fable 5.2?

The closest current models from other labs on capability, price and release date are GPT-6.2 Astra (OpenAI, ECI 174.2, $8.00 / $40.00), Gemini 4 Carbon (Google, ECI 169.8, $2.00 / $10.00), Muse Wonder (Meta, ECI 161.7, $4.00 / $20.00) and Fugu Ultra v2 (Sakana AI, $5.00 / $30.00).