PLPramaana

Anthropic

Claude Fable 5

Release date: 6/9/2026

Compare models

Anthropic's high-reliability frontier model, evaluated with extended reasoning enabled.

Pramaana IndexAccuracy: LeadingLatency: StrongCost: Standard
Accuracy / Pramaana Index
73.72%± 0.48
Latency / Pramaana Index
470.83s
Cost / Test / Pramaana Index
$2.90
Context window
1M
Max output tokens
128k
Input modality
Text · Image · File
Hyperparameter settings

temperature: 0.0

reasoning: extended

fallbacks: disabled

Benchmark performance for Claude Fable 5

BenchmarksAccuracyRanking
Pramaana Index73.72% ± 0.481 / 35
Pramaana Multimodal Index71.71% ± 0.651 / 24
Legal Research Fidelity73.32% ± 0.821 / 18
Financial Modeling Agent68.40% ± 0.991 / 22
Clinical Documentation Safety69.52% ± 1.161 / 19
MATH-50081.52% ± 1.332 / 112
GPQA Diamond87.60% ± 1.502 / 121
MMLU-Pro83.02% ± 1.672 / 119
Adaptive Tutoring Dialogue69.32% ± 0.481 / 16
SWE-bench Verified62.30% ± 0.652 / 69
LiveCodeBench79.21% ± 0.822 / 127
Agentic Reliability59.91% ± 0.991 / 12
Strategic Games Arena68.00% ± 1.161 / 20
Benefits Navigator71.30% ± 1.331 / 16

License: Proprietary · Evaluation settings, model endpoints, and benchmark versions are frozen with each published run.

Read our methodology