readable AI benchmarks

IBM

Granite 4.2 8B

reasoning · open weights · Aug 25, 2026

Granite 4.2 8B has a Quality Score of 20.2, ranking 126th among 262 scored models. Per 1M tokens, pricing is $0.060 input, $0.250 output, and $0.015 cached input. It accepts text input, outputs text, and has a 131,072 token context window. Its Reliability Score is 41.4, ranking 85th among 262 scored models, above average (total average: 33.3). Its Value Score is 20.8, low compared with the total Quality average of 26.7.

Quality Score20.2
Value Score20.8
Reliability41.4
Cache Discount75%

Model specification

Reasoning
reasoning
Input modalities
text
Output modalities
text
Context window
131,072 tokens
Weights
open weights

Token prices USD per 1M tokens

Input
$0.060
Output
$0.250
Cached Input
$0.015
Cached Output
not available
Cost per task
$0.016

Capability

Intelligence
19.6
Coding
22.4
Agentic
9.2
Omniscience
-17.2
Correct
11.2%
Blended price
$0.048

Answer outcomes

Correct 11.2%Incorrect 28.4%Abstained 60.4%

Artificial Analysis benchmarks

GDPval-AA v2
10%
τ³-Banking
8%
Terminal-Bench v2.1
18%
SciCode
30%
Humanity’s Last Exam
10%
GPQA Diamond
63%
CritPt
0%
AA-Omniscience
41%
AA-LCR
43%

Similar models