AI BENCHY
Advertise here
#157

Mercury 2.5 Preview

Inception Release: 2026-09-02 Tested on: 2026-09-02 02:06 inception/mercury-2.5-preview::high
~100BOtherClosedEstimated
(high) (medium) (low) (none)

Summary

Mercury 2.5 Preview scores 6.6 on AI BENCHY and ranks #157. It has 10.0 reliability, a 65.2% pass rate, $0.030 total cost, and 4.01s average response time.

What makes Mercury 2.5 Preview unique: It stands out most in General Intelligence, where it ranks #4, while Domain specific is its weakest area at #13.

Model facts

Researched on 2026-09-02

Estimated
Parameters
~100B
Architecture
Other
Availability
Closed
License
-

Best estimate from public evidence; the vendor did not disclose every value. Inception identifies Mercury 2.5 as a diffusion LLM but does not disclose its parameter count; the estimate follows the Mercury 2 family record.

Parameter source Availability source

Score

6.6

Consistency

8.2

Reliability

10.0

Total Cost (Current Price)

$0.030

Total Output Tokens

163,505

Total Input Tokens

115,759

Input Price

$0.040 / 1M

Output Price

$0.150 / 1M

Tests Correct

Wrong Tests: 10

Attempt pass rate: 65.2%

Flaky tests

5

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

4.01s

Response Time (max): 17.71s

Response Time (total): 88.25s

Wrong answer: 6 Did not follow instructions: 2 API error: 1 Invalid tool call: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#157 Mercury 2.5 Preview

high
Cost
$0.001
Time
4.5s
Tokens
3,851 tok

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Mercury 2.5 PreviewhighvsQwen3.6 35B A3BmediumMercury 2.5 PreviewhighvsQwen3.5 Plus 2026-02-15noneMercury 2.5 PreviewhighvsGemini 3.1 Flash LitelowMercury 2.5 PreviewhighvsQwen3.6 27BmediumMercury 2.5 PreviewhighvsInkling SmallmediumMercury 2.5 PreviewhighvsLaguna XS 2.1mediumFree AvailableMercury 2.5 PreviewhighvsGemini 3.1 Flash Lite PreviewlowMercury 2.5 PreviewhighvsGemini 3.1 Flash Lite PreviewnoneMercury 2.5 PreviewhighvsQwen3.5-27BnoneMercury 2.5 PreviewhighvsQwen3.6 Max Previewnone

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 10.0 10.0
Coding 6.7 7.8
Combined 3.8 1.6
Data parsing and extraction 10.0 10.0
Domain specific 2.9 4.4
General Intelligence 5.1 10.0
Instructions following 9.8 10.0
Puzzle Solving 10.0 10.0
Tool Calling 3.0 10.0
Trivia 3.0 10.0

Compared models