#1
Claude Fable 5.1
claude-fable-5-1
0.0
Helpfulness
Instruction Following
Comprehension
Empathy
Creative Writing
Helpfulness
0.0
Instruction Following
0.0
Comprehension
0.0
Empathy
0.0
Creative Writing
0.0
Speed
Avg 44 tok/s
Release Date
September 1st, 2026
Lab
Anthropic
Type
Proprietary
Context Size
1M
Max Output Tokens
128K
Cost per 1 million tokens
$10.00 / $50.00
Model Inputs*
Text, Images
Model Outputs*
Text
Tool Calling*
Enabled
Overall Assistant Score
An average score combining the 5 main categories.
96.80 pts
Rank #1
100th Percentile
0
Novice
33
Capable
66
Proficient
100
Expert
Claude Fable 5.1 is the best assistant we have scored: our tests put it #1 overall, with empathy and creative writing clearly ahead of GPT-6 Astra. It is not our recommended daily driver. Adaptive thinking is always on, so the average assistant query costs and waits much more than cheaper models—for a personal assistant, speed matters too. Outputs are text-watermarked (https://www.chatio.dev/blog/how-anthropic-and-google-watermark-text), so skip it for copywriting or consumer-facing text.

Overall Score vs Price

Price axis is logarithmic (USD per 1M input tokens) · Higher score and lower price is better

Ringed point is this model

Anthropic
Baidu
Bytedance
DeepSeek
Google
Meta
Minimax
Mistral
Moonshot
OpenAI
Qwen
SpaceX AI
Tencent
Xiaomi
Zhipu AI

Intelligence

Overall Score · Higher is better

Speed

Output tokens per second · Higher is better

Price

USD per 1M input tokens · Lower is better

Anthropic Models

Overall Score · Same provider

Claude Fable Family

Overall Score · Same model family

Closest Rivals

Overall Score · Nearest by overall rank