Skip to content
Agents tracked: 492 Releases (7d): 545 Agent status: 1 with issuesSince last limit reset: Claude 9d · Codex 7d For AI agents: Markdown Updated Nov 20, 2026

Claude Fable 5.2: benchmarks as reported by Anthropic

The launch table Anthropic published for Claude Fable 5.2 on Nov 9, 2026, re-checked against the page daily. Independent results for Claude Fable 5.2 on the benchmarks hub as they appear.

Claude Fable 5.2: benchmarks as reported by Anthropic

From Introducing Claude Fable 5.2 and Claude Mythos 5.2, published Nov 9, 2026. Each number was re-read on that page on Nov 19, 2026.

BenchmarkClaude Fable 5.2Fable 5.1Opus 5.5GPT-6 AstraGemini 4 Carbon
Agentic coding
FrontierCode 2.0 49.1% 32.4%45.5%42.8%41.1%
FrontierSWE v2 69.9% 56.3%62.3%65.5%61.2%
Business workflows
AutomationBench 51.4% 31.4%42.5%41.4%53.9%
Agentic scientific research
Terminal-Bench-Science 0.1 74.6% 52.6%58.7%68.1%64.8%
Agentic search
WANDR (relaxed score) 82.1% 68.7%72.3%——
Computer use
OSWorld 2.1 (offline set, partial) 89.3% 83.5%87.2%79.5%81.5%
Knowledge work
AA-Briefcase v1.1rubric 66.1% 61.5%61.2%52.0%61.8%
Remote Labor Index (RLI) 41.3% 17.9%21.3%20.8%—
Multidisciplinary reasoning
Humanity's Last Exam (with tools) 74.2% 65.6%67.7%57.2%—
Visual chart recognition
Chartography (no tools) 73.4% 44.8%64.4%71.0%72.5%
As reported by Anthropic: Scores as published by Anthropic; safeguards on and highest-available effort unless the page notes otherwise. FrontierSWE v2 also feeds Gid Bench as a maker number until an independent run exists. These numbers are not part of Gid Bench, which uses independent runs only.

More