Granite 4.2 3B
By IBM · released 2026-08-25 (2 weeks) · snapshot 2026-09-09
- Intelligence index
- 9.10
- Cost to run the AA index
- $14.9
- Price per 1M tokens (in / out)
- $0.03 / $0.12
- End-to-end time
- 12.1s
Is Granite 4.2 3B on the Pareto frontier, and what beats it?
On the Pareto frontier
Yes. Nothing live under $10,000 is both smarter and cheaper to run.
The daily check URL below returns 409 the day that changes.
Open in the picker Daily check URL: 200 on the frontier, 409 once beaten
Granite 4.2 3B today
Intelligence: Artificial Analysis Intelligence Index. Cost: total cost to run that benchmark, so reasoning tokens count. API id: granite-4-2-3b.
Compare head to head
Suggested: what beats it, its effort levels, and frontier models near its price.
What the Granite 4.2 3B release did to the frontier on 2026-08-25
1 of 1 effort levels reached the frontier at launch (Granite 4.2 3B). Impact 2.17, #13 of 28 releases this past year.
Should you use Granite 4.2 3B?
Should you use Granite 4.2 3B?
Yes, for the money: nothing live under $10,000 is both smarter and cheaper to run, so it sits on the value frontier.
What is better and cheaper than Granite 4.2 3B?
Nothing: no live model under $10,000 is both smarter and cheaper.
How much does Granite 4.2 3B cost?
$0.03/$0.12 in/out per 1M tokens, and $14.9 to run the full Artificial Analysis Intelligence Index, the honest all-in number for reasoning models.
Is Granite 4.2 3B on the Pareto frontier?
Yes: Granite 4.2 3B is on the intelligence-vs-cost Pareto frontier.
Frontier membership by axis pair
Other axis pairs where Granite 4.2 3B is Pareto-optimal.
| Y axis | X axis | On the frontier |
|---|---|---|
| AA intelligence index | AA total cost to run (USD) | |
| AA intelligence index | End-to-end total seconds | |
| AA intelligence index | Output speed tokens/sec |
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/models/granite-4-2-3b · Markdown: /api/v1/models/granite-4-2-3b?format=md · llms.txt · agent skill