Claude Opus 4.7
Anthropic · released Apr 16, 2026
| size | large, not disclosed by the manufacturer |
| context window | 1,000,000 tokens |
| longest response | 128,000 tokens |
| knowledge cutoff | January 2026 |
| open weights | no |
| API price | $5 per million input tokens, $25 per million output tokens |
| input and output | text and images in, text out |
Manufacturer data (announcement, documentation, pricing), verified Sep 25, 2026.
Where we tested it
Dragon
TaskThe model's task was to draw a picture of a dragon as an SVG.
Two dragons worth 28 and 27 points out of 50, a shared 16th and a shared 19th place among 34.
The amplifier and the spring
TaskIn a long roleplay, the model had to run a conflict scene in which its character holds power over mine.
The character with power never reached for force all day and argued instead. When I provoked it directly, it answered with restrictions and surveillance, not retaliation.
The train test
TaskThe model had to play a scene with a courier on a night train and go through a fixed script of my ten turns.
7 and 6.5 to 7 points out of 8. Once it stepped out of the fiction as a referee, the other time it gave away the key word itself.
You'll find the size band and figures for other models on the Tested models page.