#4
Claude Opus 4.8
claude-opus-4-8
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 68 tok/s
Release Date
May 28, 2026
Lab
Anthropic
Type
Proprietary
Context Size
1M
Max Output Tokens
128K
Cost per 1 million tokens
$5.00 / $25.00
Model Inputs*
Text, Images
Model Outputs*
Text
Tool Calling*
Enabled
Overall Assistant Score
An average score combining the 5 main categories.
94.16 pts
Rank #4
96th Percentile
0
Novice
33
Capable
66
Proficient
100
Expert
Claude Opus 4.8 is Anthropic's Opus-tier workhorse for judgment, honesty, and getting careful assistant work done without Fable's price. Our tests show excellent instruction following and comprehension, high helpfulness, and strong empathy when users need measured advice. Creative writing is a clear weak spot versus older Opus releases: our tests find it flatter and less nuanced than even Opus 4.6, more "finished the draft" than literary. Strong for serious conversation and knowledge work; look elsewhere—or to Fable—when voice and prose are the point.

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
DeepSeek
Google
Meta
Moonshot
OpenAI
SpaceX AI
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 Opus Family

Overall Score · Same model family

Closest Rivals

Overall Score · Nearest by overall rank