Models · Anthropic

Claude Fable 5

flagship Closed 2026-06-09

Lab: Anthropic
Released: 2026-06-09
Parameters: unknown
Context window: 1M
Input: text, image
Output: text
Weights: Closed
Official page: platform.claude.com ↗

Benchmark scores

DeepSWE agentic 69.7%

69.7% 2026-08-22 GLM-5.3 matches Claude Fable 5 accuracy on DeepSWE at a fifth of the cost
69.7% 2026-08-18 DeepSeek V4 Pro 0813 vs Claude Fable 5 on DeepSWE: Cost, Coding, and Routing

HarmBench safety 6.1%

6.1% 2026-06-17 Red-Team Study Finds Frontier LLMs Remain Vulnerable to Automated Attacks
6.1% 2026-06-17 Red-Team Study Finds Frontier LLMs Remain Vulnerable to Adaptive Attacks

Humanity's Last Exam reasoning 53%

53% 2026-06-09 Claude Fable 5 launches at #1 on the Artificial Analysis Intelligence Index

LMSYS Arena (Elo) general 1760 Elo

1760 Elo 2026-07-24 Kimi K3 launches as largest open-weight model; Muse Spark 1.1 released

Related articles