Advertise here
#312

Mercury 2.5

Inception Release: 2026-09-08 Tested on: 2026-09-08 22:51 inception/mercury-2.5::none
~100BOtherClosedEstimated
(high) (medium) (low) (none)

Summary

Mercury 2.5 scores 3.9 on AI BENCHY and ranks #312. It has 10.0 reliability, a 25.8% pass rate, $0.016 total cost, and 2.45s average response time.

What makes Mercury 2.5 unique: It stands out most in Domain specific, where it ranks #1, while Anti-AI Tricks is its weakest area at #10.

Model facts

Researched on 2026-09-08

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 or topology. The low-confidence 100B estimate follows the existing Mercury family records; it is not a vendor-reported count.

Parameter source Availability source

Score

3.9

Consistency

7.5

Reliability

10.0

Total Cost (Current Price)

$0.016

Total Output Tokens

71,143

Total Input Tokens

123,740

Input Price

$0.040 / 1M

Output Price

$0.150 / 1M

Tests Correct

Wrong Tests: 20

Attempt pass rate: 25.8%

Flaky tests

7

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

Response Time (avg)

2.45s

Response Time (max): 36.48s

Response Time (total): 53.83s

Wrong answer: 14 Did not follow instructions: 4 Extra formatting: 1 Invalid tool call: 1

Hamster playing table tennis

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

#312 Mercury 2.5

none
Cost
$0.001
Time
2.7s
Tokens
2,749 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 2.5nonevsNemotron 3.5 LightningnoneFree AvailableMercury 2.5nonevsLing 3.0 TinyhighMercury 2.5nonevsGranite 4.1 8BnoneMercury 2.5nonevsLing 3.0 TinylowMercury 2.5nonevsLing 3.0 TinynoneMercury 2.5nonevsQwen3.5-9BmediumMercury 2.5nonevsGrok 4.20noneMercury 2.5nonevsLing 3.0 TinymediumMercury 2.5nonevsGranite 4.2 8BnoneMercury 2.5nonevsgpt-oss-120bnoneFree Available

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 3.8 8.3
Coding 3.7 7.0
Combined 3.0 10.0
Data parsing and extraction 2.8 1.7
Domain specific 5.3 7.2
General Intelligence 4.3 9.9
Instructions following 4.9 6.9
Puzzle Solving 3.7 7.2
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models