readable AI benchmarks

IBM

Granite 4.2 3B

reasoning · open weights · Aug 25, 2026

Granite 4.2 3B has a Quality Score of 16.1, ranking 153rd among 262 scored models. Per 1M tokens, pricing is $0.030 input, $0.120 output, and $0.0075 cached input. It accepts text input, outputs text, and has a 131,072 token context window. Its Reliability Score is 42.6, ranking 81st among 262 scored models, above average (total average: 33.3). Its Value Score is 16.0, low compared with the total Quality average of 26.7.

Quality Score16.1
Value Score16.0
Reliability42.6
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
$0.030
Output
$0.120
Cached Input
$0.0075
Cached Output
not available
Cost per task
$0.0047

Capability

Intelligence
14.3
Coding
17.5
Agentic
1.9
Omniscience
-14.7
Correct
9.2%
Blended price
$0.023

Answer outcomes

Correct 9.2%Incorrect 23.9%Abstained 66.9%

Artificial Analysis benchmarks

GDPval-AA v2
0%
τ³-Banking
6%
Terminal-Bench v2.1
14%
SciCode
25%
Humanity’s Last Exam
7%
GPQA Diamond
56%
CritPt
0%
AA-Omniscience
43%
AA-LCR
24%

Similar models