Claude Fable 5
Claude Fable 5 is a closed AI model from Anthropic with an index score of 183 (GPT-4 = 100), rank 15 of 217.
183167–199Index score#15 of 217 models
$10/ $50Price per million tokensinput / output · via OpenRouter
1MContext windowtokens it reads at once
9 Jun 2026ReleasedReplaced by Claude Fable 5.1
Where it sitson the frontier
Among the first models to reach this score.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsevidence
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 12 leaderboards
The Model Indexthird-party tests, our method#15of 217 · 183
LMArena · Visionunderstanding images#1of 145 · 1309LMArena · Codingcoding questions in chat#3of 357 · 1552LMArena · Long promptslong, complex requests#3of 340 · 1522LMArena · Creative writingcreative writing#4of 360 · 1501LMArena · Mathsmaths questions in chat#4of 346 · 1520LMArena · Textoverall chat quality, judged by people#5of 362 · 1504LMArena · Instruction followingfollowing detailed instructions#5of 362 · 1509Epoch Capabilities Indexoverall capability from many benchmarks#12of 231 · 164.2LMArena · Documentsworking with long documents#4of 40 · 1497LMArena · WebDevbuilding web apps and front ends#25of 133 · 1626LMArena · Searchanswering with web search#10of 41 · 1228Best attop three among current models
Where to use itprices checked 18 Nov 2026
API providers 5 · per 1M tokens, in / out
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released9 Jun 2026
Announcementanthropic.com ↗
Release entryClaude Fable 5 + Mythos 5 (restricted)
Scored asClaude Fable 5
Output speed–
Hugging Faceclosed weights