AnthropicClosed

Claude Fable 5.1

Name: Claude Fable 5.1
Author: Anthropic

#1 of all models across the language-model boards, #1 on 1 board.

Overall rank

57.3

Podium score

Board wins

Sep 1, 2026

Released

9Chat 2Reasoning 2Coding 5Agents and tools 5Vision ·Speed ·Price

Chat

Which model people prefer in blind head-to-head conversations.

Text Arena

1,498 · of 100

as “claude-fable-5.1-max”

Reasoning

Hard questions with checkable answers: science, math, puzzles, expert exams.

LiveBench

83.4% · of 59

as “claude-fable-5-1-max-effort”

AA Intelligence

53.4 · of 100

as “Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)”

HLE

46.5% · of 42

as “Fable 5.1 (xhigh)”

Epoch ECI

165.0 · of 100

as “Claude Fable 5.1”

ARC-AGI-2

90.0% · of 78

as “Claude Fable 5.1 (Max)”

Coding

Fixing real repositories, building web apps, editing code.

SWE-Bench Pro

92.2% · of 11

as “Fable 5.1 (Claude Code) high”

WebDev Arena

1,755 · of 100

as “claude-fable-5.1-max”

Agents and tools

Calling functions and tools, using MCP servers, searching the web.

MCP Atlas

87.2% · of 34

as “Fable 5.1”

Vision

Understanding images, charts and documents.

Document Arena

1,513 · of 38

as “claude-fable-5.1-max”

Vision Arena

1,289 · of 100

as “claude-fable-5.1-max”

Speed

Output tokens per second on the vendor API.

AA Speed

66 tok/s · of 24

as “Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)”

Price

Blended API price per million tokens, cheapest first.

AA Price

$20.0 · of 27

as “Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)”

More from Anthropic

Claude Opus 5.5 Claude Opus 5 Claude Sonnet 5 Claude Fable 5 Claude Opus 4.8 Claude Opus 4.7 Claude Sonnet 4.6 Claude Opus 4.6

Scores as published by each board on the capture date. Model names and logos belong to their owners; logos via logo.dev.