readable AI benchmarks

Inception

Mercury 2

reasoning · closed weights · Feb 20, 2026

Quality Score20.7
Value Score23.1
Reliability24.6
Cache Discount90%

Model specification

Reasoning
reasoning
Input modalities
text
Output modalities
text
Context window
128k tokens
Weights
closed weights

Token prices USD per 1M tokens

Input
$0.250
Output
$0.750
Cached Input
$0.025
Cached Output
not available
Cost per task
$0.080

Capability

Intelligence
21.9
Coding
31.1
Agentic
9.5
Omniscience
-50.7
Correct
21.2%
Blended price
$0.143

Answer outcomes

Correct 21.2%Incorrect 71.9%Abstained 6.9%

Artificial Analysis benchmarks

GDPval-AA v2
10%
τ³-Banking
9%
Terminal-Bench v2.1
27%
SciCode
39%
Humanity’s Last Exam
17%
GPQA Diamond
77%
CritPt
1%
AA-Omniscience
25%
AA-LCR
41%

Similar models