IBMReleased Aug 25, 2026
Open weights

Granite 4.2 3B

Its best results are in Programmatic Reasoning (#131 of 181) and Knowledge Work (#152 of 180); weakest in Academic Reasoning (#204 of 211). Three of seven dimensions are estimated, because no results are published yet.

Composite IQ ?
88
#168 of 180 ranked models · ahead of 7%
US (Other)'s #20 model, behind Inkling 2 Small (126)
Effective cost
$0.068/1M
$0.03 in · $0.12 out
Speed
40tok/s
64s typical response
Context
131Ktokens
131,072 tokens in
Evidence
4of 7 dims
11 of 39 benchmarks

IQ profile

Seven dimensions, averaged equally into Composite IQ
Compare shape with
100 avg human130160● Abstract82#159 · est.● Math79#174 · est.● Academic97#204● Coding99#131● Computer Use70#176 · est.● Reliability94#165● Knowledge Work92#152
MeasuredEstimated95% range
#1 in dimensionTop-10 median

IQ dimensions

Ranks are among models scored in each dimension. Ranges are model-based estimates, not calibrated confidence intervals.

DimensionIQPosition · 70–160RankEvidencevs Top 10
82estimated
#159 / 209
0/5
-50
79estimated
#174 / 180
0/5
-68
97measured
#204 / 211
4/6
-38
99measured
#131 / 181
1/6
-51
70estimated
#176 / 180
0/5
-74.5
94measured
#165 / 181
2/6
-40.5
92measured
#152 / 180
4/6
-40
Granite 4.2 3B95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

11 of 39 Composite IQ benchmarks have a published result for Granite 4.2 3B. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 82 · estimatedNo published results0/5
No published results yet for Abstract Reasoning. The 82 shown is estimated from Granite 4.2 3B's other dimensions and will be replaced as results come in. How estimates work
ARC-AGI-3ARC-AGI-2ARC-AGI-1Chess PuzzlesMystery Game Puzzles
Mathematical Reasoning IQ 79 · estimatedNo published results0/5
No published results yet for Mathematical Reasoning. The 79 shown is estimated from Granite 4.2 3B's other dimensions and will be replaced as results come in. How estimates work
FrontierMath Tier 4FrontierMath Tier 1-3ProofBench 1.1MathArenaAIME
Academic Reasoning IQ 97 · measuredBest: #144 of 159 on CritPt4/6
BenchmarkScorePercentileRankLeader · source
CritPt
Artificial Analysis v4.3.2 · Oct 27
0.0%
#144 / 159
Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 27
Humanity's Last Exam
Artificial Analysis v4.3.2 · Oct 27
6.6%
#153 / 167
Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 27
GPQA Diamond
Artificial Analysis v5.0.1 · Nov 13
55.9%
#139 / 147
Leader gpt-6-astra · 96.1%
Artificial Analysis v5.0.1 · Nov 13
SciCode
Artificial Analysis v4.3.2 · Oct 27
25.3%
#161 / 167
Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
MMLU-ProMMMU-Pro
Programmatic Reasoning IQ 99 · measuredBest: #84 of 109 on Terminal-Bench 4.01/6
BenchmarkScorePercentileRankLeader · source
Terminal-Bench 4.0
Artificial Analysis v5.0.1 · Nov 13
0.0%
#84 / 109
Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v5.0.1 · Nov 13
Not yet reported
LiveCodeBenchIOI V2ProgramBenchSWE-rebenchMirrorCode
Computer Use IQ 70 · estimatedNo published results0/5
No published results yet for Computer Use. The 70 shown is estimated from Granite 4.2 3B's other dimensions and will be replaced as results come in. How estimates work
BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam
Reliability IQ 94 · measuredBest: #102 of 158 on AA Omniscience2/6
BenchmarkScorePercentileRankLeader · source
AA Omniscience
Artificial Analysis v4.3.2 · Oct 27
-14.7
#102 / 158
Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 27
AA Long Context Reasoning v1.1
Artificial Analysis v4.3.2 · Oct 27
24.3%
#151 / 154
Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 27
Not yet reported
SimpleQA VerifiedBullshitBench v2IFBenchFACTS Grounding
Knowledge Work IQ 92 · measuredBest: #109 of 124 on GDPval-AA v2.1 normalized score4/6
BenchmarkScorePercentileRankLeader · source
GDPval-AA v2.1 normalized score
Artificial Analysis v4.3.2 · Oct 23
0.0%
#109 / 124
Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 23
AA-Briefcase v1.1 Elo
Artificial Analysis v5.0.1 · Nov 13
98.8
#111 / 116
Leader sonnet-5.5 · 1,816
Artificial Analysis v5.0.1 · Nov 13
AutomationBench-AA
Artificial Analysis v5.0.1 · Nov 13
0.4%
#109 / 112
Leader Gemini 4 Carbon · 77.8%
Artificial Analysis v5.0.1 · Nov 13
GDP.pdf
Artificial Analysis v5.0.1 · Nov 13
0.2%
#111 / 111
Leader Muse Wonder Full · 32.6%
Artificial Analysis v5.0.1 · Nov 13
Not yet reported
AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released
Developer
IBM
Access
Apache 2.0 open weights on Hugging Face for local and edge deployment
Weights
Open weightsDownloadable and self-hostable
Inputs
text
Context
131,072 tokens
API price
$0.03 in · $0.12 outper 1M tokens · full pricing

What IBM says

  • Granite 4.2 3B is IBM's compact dense reasoning model for edge and resource-constrained deployments.
  • It supports a native 128K context window, reasoning-augmented tool calling, and switchable full, low-effort, and non-thinking modes.

Limitations

  • IBM says multilingual performance can vary relative to English.
  • The model may produce inaccurate, biased, or unsafe responses, and IBM recommends deployment alongside additional safety controls.

Publisher statements, not AI IQ measurements.

About this score

How much of Granite 4.2 3B's IQ is measured, and how the rest is estimated.

11/ 39benchmark results
4/ 7dimensions measured
3estimated
0/5Abstract
0/5Math
4/6Academic
1/6Coding
0/5Computer Use
2/6Reliability
4/6Knowledge Work
Bar height = share of benchmarks with resultsEstimatedNo score
Measured results are shown as published

Every score in the benchmark tables comes from its source unchanged. Composite IQ averages all seven dimensions equally.

Estimates fill the gaps

Abstract Reasoning, Mathematical Reasoning and Computer Use have no published results yet, so their IQ is estimated from Granite 4.2 3B's other dimensions.

A missing result is not a zero

Benchmark versions and evaluation harnesses can differ between sources, so small gaps between models may not be meaningful.

Read the scoring methodology →

Compare Granite 4.2 3B

Side-by-side pages for the models it is most often weighed against.

Granite 4.2 3BComparison model
Browse all comparisons →

Frequently asked questions

Quick answers about Granite 4.2 3B, drawn from the same data as the rest of this page.

How AI IQ scores models →Get this data from the API →
What is the IQ of Granite 4.2 3B?
Granite 4.2 3B has an estimated IQ of 88, which ranks #168 out of 180 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
What is Granite 4.2 3B best at?
Among the dimensions with published benchmark results, Granite 4.2 3B ranks highest in Programmatic Reasoning (IQ 99, #131 of 181) and Knowledge Work (IQ 92, #152 of 180). Its lowest-ranked measured dimension is Academic Reasoning (IQ 97, #204 of 211).
What is the EQ of Granite 4.2 3B?
Granite 4.2 3B doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for Granite 4.2 3B. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Granite 4.2 3B been tested on?
Granite 4.2 3B has published results on 11 of the 39 benchmarks behind AI IQ's Composite IQ, including AA Omniscience, Terminal-Bench 4.0, GDPval-AA v2.1 normalized score, CritPt and Humanity's Last Exam.
How much does Granite 4.2 3B cost?
Granite 4.2 3B costs $0.03 per million input tokens and $0.12 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.0675 per million input-plus-output tokens.
Is Granite 4.2 3B open source?
Yes — Granite 4.2 3B is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed Granite 4.2 3B?
Granite 4.2 3B was developed by IBM, based in the United States.
When was Granite 4.2 3B released?
Granite 4.2 3B was publicly released on August 25, 2026.
How large is the context window of Granite 4.2 3B?
Granite 4.2 3B supports a context window of 131,072 tokens — the amount of text it can consider in a single request.
How fast is Granite 4.2 3B?
In independent measurements, Granite 4.2 3B generates about 40 tokens per second, with a typical response time of about 64 seconds.