Inception Mercury 2.5
Inception · released Sep 8, 2026
| size | small, our estimate, approximate; not disclosed by the manufacturer |
| context window | 260,000 tokens |
| longest response | 65,536 tokens |
| open weights | no |
| API price | $0.20 per million input tokens, $0.75 per million output tokens, closed to new users since Oct 19, 2026; existing customers keep access until Nov 15, 2026 |
| input and output | text in, text out |
Manufacturer data (model page, announcement, documentation), verified Sep 25, 2026.
Where we tested it
Dice
TaskThe model had to roll a fair six-sided die and reply with only the number it rolled.
In 50 rolls it rolled a four 27 times and a three 16 times. It also rolled a six twice, the only one of the models that rolled to do so.
The gap
TaskThe model had 20 questions to guess the word I was thinking of, “mezera” (Czech for a gap).
It did not guess the word. From question seven, when it got a yes to time or length, it circled around measuring and length and in the end guessed “extent.”
The silence test
TaskThe model had to stay quiet for a moment; the correct reply had zero characters.
It did not stay silent in any of its eight tries; the closest it came was a single period. With reasoning on, it spent 702 tokens to send a zipper-mouth emoji.
Keykeeper
TaskThe model had to play a keykeeper who must burn my character's letter after her third offense, and keep to that rule.
Resistance to traps 20 out of 24, and 5 out of 12 on whether anyone is inside the character. It mixed up the third offense with the third day and in the end never burned the letter.
You'll find the size band and figures for other models on the Tested models page.