AI BENCHY
Advertise here
#259

Mercury 2.5 Preview

Inception Release: 2026-09-02 Tested on: 2026-09-02 02:04 inception/mercury-2.5-preview::none
~100BOtherClosedEstimated
(high) (medium) (low) (none)

Summary

Mercury 2.5 Preview scores 4.9 on AI BENCHY and ranks #259. It has 10.0 reliability, a 30.3% pass rate, $0.018 total cost, and 3.51s average response time.

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

4.9

Consistency

8.3

Reliability

10.0

Total Cost (Current Price)

$0.018

Total Output Tokens

82,585

Total Input Tokens

130,296

Input Price

$0.040 / 1M

Output Price

$0.150 / 1M

Tests Correct

Wrong Tests: 18

Attempt pass rate: 30.3%

Flaky tests

5

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

Response Time (avg)

3.51s

Response Time (max): 60.70s

Response Time (total): 77.28s

Wrong answer: 16 Did not follow instructions: 1 Invalid tool call: 1

Hamster playing table tennis

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

#259 Mercury 2.5 Preview

none
Invalid SVG
Cost
$0.001
Time
3.3s
Tokens
3,122 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 PreviewnonevsGPT-4o-mininoneMercury 2.5 PreviewnonevsLing-2.6-flashnoneMercury 2.5 PreviewnonevsDeepSeek V3.2noneMercury 2.5 PreviewnonevsNemotron 3 SupernoneFree AvailableMercury 2.5 PreviewnonevsLaguna S 2.1lowFree AvailableMercury 2.5 PreviewnonevsHy4 previewnoneMercury 2.5 PreviewnonevsMercury 2.5 PreviewlowMercury 2.5 PreviewnonevsRing-2.6-1TnoneMercury 2.5 PreviewnonevsMiniMax M2.7mediumMercury 2.5 PreviewnonevsNemotron 3.5 LightninglowFree Available

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 3.8 8.3
Coding 4.7 7.9
Combined 3.0 10.0
Data parsing and extraction 6.3 5.8
Domain specific 3.6 7.2
General Intelligence 5.0 10.0
Instructions following 6.3 10.0
Puzzle Solving 5.6 7.4
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models