Models / Anthropic

Claude Fable 5.2

Anthropic · Proprietary

Recent snapshotData snapshot 2026-11-12 20:04 UTCcf-1.3

Cheese Index76.7cf-1.3

Input / output$8 / $40USD / 1M tokens

Measured dimensions7 / 13Evidence

Context window1,000,000tokens · conditions unverified

Decision brief

Start by investigating Knowledge and Mathematics workloads. These are observed strengths in the dataset; validate them on your own tasks.

Coverage counts measured dimensions, not confidence. Benchmark conditions and public source links may be missing. Missing: Instructions, Multilingual, swe, terminal, gpqa, mcp.

View evidence ↓

Measured task dimensions

Coding

89.1

Agent workflows

90.0

Mathematics

93.8

Knowledge

98.0

Instructions

—

Multimodal

89.6

Multilingual

—

Reasoning

90.5

swe

—

terminal

—

gpqa

—

hle

73.2

mcp

—

Normalized dimensions. A dash means unknown, not zero.

Input / output

$8 / $40 · USD / 1M tokens

Listed dataset prices; provider terms and dates need verification. Zero without explicit free-price evidence is treated as unknown.

Cache, batch, tool fees and self-hosting costs: not verified.

Index variability stability

99.1 · 4 days of history

short_band

This is index variability, not API uptime or reliability.

Before deployment

Confirm provider endpoint, license and version.
Test your prompts, tool calls and failure recovery.
Verify prices and latency with real usage.

Evidence

These are source categories published by the dataset. Original public links, sample sizes and harness details remain incomplete.

Source category	Provenance type	As of	Original public link
Coding	core / official_json	2026-11-13	Not published
Agent workflows	core / official_json	2026-11-13	Not published
Long workflows	horizon / official_json	2026-11-13	Not published
ops	cross / official_json	2026-11-13	Not published
depth	core+cross / official_json	2026-11-13	Not published

Published index inputs · cf-1.3

Coding69.5

Agent workflows90.0

Long workflows69.9

Overall—

The method includes overall, but this snapshot does not expose that input. The total cannot be fully reproduced from these displayed axes.

Method: median of available axes, with an 8% Ops blend when present. Source scores are displayed without recalculation or imputation.

Method →

Index history

As of	Cheese Index
2026-11-10	77.1
2026-11-11	76.8
2026-11-12	76.8
2026-11-13	76.7

Only recorded dates are shown. Missing days are not filled.