AI BENCHY
AD
Track all your projects in one dashboard. Get 📊stats, 🔥heatmaps and 👀recordings in one self-hosted dashboard.
uxwizz.com
#276

Granite 4.2 8B

IBM Granite Release: 2026-09-02 Tested on: 2026-09-02 02:15 ibm-granite/granite-4.2-8b::high
8BDenseOpen source

Summary

Granite 4.2 8B scores 4.6 on AI BENCHY and ranks #276. It has 7.6 reliability, a 39.4% pass rate, $0.084 total cost, and 235.89s average response time.

What makes Granite 4.2 8B unique: It stands out most in Combined, where it ranks #1, while Data parsing and extraction is its weakest area at #10. It uses unusually many reasoning tokens, which can help explain its slower or more expensive runs.

Model facts

Researched on 2026-09-02

Reported
Parameters
8B
Architecture
Dense
Availability
Open source
License
Apache-2.0

IBM publishes the weights, runnable implementation guidance, and training methodology under Apache-2.0.

Score

4.6

Consistency

6.3

Total Output Tokens

612,215

Total Input Tokens

43,278

Input Price

$0.100 / 1M

Output Price

$0.150 / 1M

Tests Correct

Wrong Tests: 18

Attempt pass rate: 39.4%

Flaky tests

10

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

Response Time (avg)

235.89s

Response Time (max): 685.87s

Response Time (total): 5189.64s

Hamster playing table tennis

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

#276 IBM: Granite 4.2 8B

high
Provider returned error
Cost
$0.000
Time
0.2s
Tokens
0 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

Category Breakdown

Category Score Consistency Tests Correct
Anti-AI Tricks 6.9 5.8
Coding 3.8 4.3
Combined 5.0 6.8
Data parsing and extraction 3.8 5.8
Domain specific 2.9 7.2
General Intelligence 5.5 10.0
Instructions following 6.4 5.8
Puzzle Solving 2.9 4.4
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models