Skip to content
Leaderboard / Anthropic

Claude Fable 5

Anthropic High CLOSED Proprietary
Verdict

Claude Fable 5 is a strong all-rounder: #21 of 406 by SIscope score. Strongest at following instructions, weaker at reasoning & knowledge. On the pricey side: costs more than 99% of priced models.

SIscope score
89
#21 of 406
Human-vote rank
#11
of 419 · score 1492
Price per 1M tokens, blended
$20
$10 in · $50 out
Context window
1M
1,000,000 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding90
Agents & tools84
Reasoning & knowledge81
Math97
Chat quality98
Following instructions98

What goes into each category · Compare with DeepSeek V4.1 Flash

Human-vote score over time

Text arena, recent updates.

148514931500Oct 15: 1491.2 (rank #10)Oct 20: 1491.3 (rank #10)Oct 22: 1490.9 (rank #12)Oct 28: 1491.6 (rank #11)Oct 15Oct 28

Coding votes over time

WebDev arena, recent updates.

161516231630Oct 16: 1624.7 (rank #21)Oct 18: 1624.5 (rank #22)Oct 19: 1624.3 (rank #23)Oct 22: 1624.1 (rank #23)Oct 23: 1624.3 (rank #24)Oct 27: 1623.9 (rank #25)Oct 28: 1624.1 (rank #26)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond83%
PhD-level science multiple choice questions
Better than 64% of 333 tested models. Run Aug 6, 2026.
OTIS Mock AIME 2024-25100%
Competition maths problems in the style of AIME
Better than 96% of 313 tested models. Run Aug 6, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#11 of 41914921487–149645,341Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#26 of 15216241618–163016,951Oct 28, 2026
Vision arena
Questions about images, blind human votes
#1 of 16713251318–133214,714Oct 19, 2026
Agent arena
Real multi-step agent sessions
#10 of 577.8%6.7–8.9%51,112Nov 2, 2026

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$10
per 1M tokens
Output
$50
per 1M tokens
Blended
$20
3 in : 1 out, per 1M
Context window
1M
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (anthropic/claude-fable-5), Nov 4, 2026. Providers can charge differently.