Skip to content
VECTOR WIREAI INTELLIGENCE
⌘K
UTC
Models790Deals30d240Policy30d70, 15 pendingSources1,102

Granite 4.2 3B

IBMReleased Aug 25, 2026Current

Score basis

Granite 4.2 3B is capable in instruction following and behind the leaders in factuality and agentic tasks. Too few results yet to rate reasoning, coding, safety, long context, math, multimodal tasks, or multilingual tasks.

Capability profile

Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.

Capable1 capability
  1. Instruction Following−24.8%1 of 3
Limited2 capabilities
  1. Factuality−36.3%2 of 4
  2. Agentic−56.9%3 of 7
Not rated7 capabilities

Too few results yet to rate Reasoning, Coding, Safety, Long Context, Math, Multimodal or Multilingual.

Price

$0.03input$0.12outputper million tokens

From Artificial Analysis · All prices

Cheaper than 97% of 322 priced models · 3:1 input-to-output blend, log scale

Evidence

32results on29benchmarks

  • 1 independently verified
  • 15 aggregator
  • 16 vendor-reported

From 4 sources · latest Oct 24, 2026 · How verification works

Research

1 paper reference Granite 4.2 3B

Granite 4.2 3B benchmark results

32 results on 29 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.

Instruction Following

CapableFull instruction following ranking

24.8% behind the leader1 of 3 ranked benchmarks measured

Factuality

LimitedFull factuality ranking

36.3% behind the leader2 of 4 ranked benchmarks measured

Agentic

LimitedFull agentic ranking

56.9% behind the leader3 of 7 ranked benchmarks measured

Reasoning

Not enough dataFull reasoning ranking

0 of 6 ranked benchmarks measured

Show 1 more reasoning resultHide 1 reasoning result

Coding

Not enough dataFull coding ranking

0 of 10 ranked benchmarks measured

Show 1 more coding resultHide 1 coding result

Long Context

Not enough dataFull long context ranking

0 of 3 ranked benchmarks measured

More results

Benchmarks outside the capability baskets. They are not ranked against a leader.

Show 11 more resultsHide 11 results

Granite 4.2 3B: common questions

Who makes Granite 4.2 3B?

Granite 4.2 3B is made by IBM.

When was Granite 4.2 3B released?

Granite 4.2 3B was released on Aug 25, 2026, according to Artificial Analysis.

What is Granite 4.2 3B good at?

Granite 4.2 3B is capable in instruction following and behind the leaders in factuality and agentic tasks. Too few results yet to rate reasoning, coding, safety, long context, math, multimodal tasks, or multilingual tasks.

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, according to Artificial Analysis. At a mix of three input tokens to one output token, it is cheaper than 97% of the 322 priced models we track.

How many benchmarks has Granite 4.2 3B been tested on?

We track 32 results for Granite 4.2 3B on 29 benchmarks from 4 sources, 1 of them independently verified. The latest was recorded on Oct 24, 2026.

About this record

Where Granite 4.2 3B's numbers come from, and every name it appears under.

Tracked since
Aug 25, 2026
Newest source mention
Aug 27, 2026

Where the results come from

Verification: 32 scores · 1 independently verified · 15 aggregator-attributed · 16 vendor-reported. How these tiers are assigned

From 4 sources on 3 sites. Artificial Analysis supplies 16 of them; the 1 independently verified result comes from 1 site. Bars are coloured by trust tier.

  • artificialanalysis.ai16
  • api.llm-stats.com12
  • huggingface.co4

Also known as

granite 4.2 3b densegranite-4-2-3b