Granite 4.2 8B
IBM · released Aug 7, 2026 · ibm-granite/granite-4.2-8b
- Type
- Open weightsApache License 2.0
- Params
- 8.8B
- Context
- 131K
about 98K words of context
Our take
Written Sep 30, 2026Granite 4.2 8B is a downloadable text model from IBM whose licence allows commercial use, changes and redistribution (Apache License 2.0). On the human-preference boards it sits near the bottom of the field, so pick it for the licence and the download rather than for measured quality.
Choose it for internal tooling, fine-tuning, or a product you intend to ship, where a permissive licence and a download matter more than leaderboard standing. It also takes long documents without your having to split them up first. Skip it if you need a model that ranks well on human-preference chat, writing or coding boards, because this one sits near the bottom of every board we hold.
The case for it
- The licence allows commercial use, changes and redistribution (Apache License 2.0), so a product built on it does not need a different licence negotiated first.
- You can download it and run it yourself rather than rent it, because the weights are published.
- The request capacity is large enough to hold a long document alongside the question, though reliable recall across all of it is unverified in our data.
The case against it
- Ranked 153rd of 168 on Arena Text (overall) as of 25 Sep 2026 and 162nd of 168 on Arena Creative Writing as of 25 Sep 2026 — these boards record which answer people preferred, not whether it was correct.
- Every score we hold comes from a human-preference arena; nothing supplied measures accuracy on reasoning, retrieval or code correctness, so those need a trial on work you can check yourself.
How good is it?
An open text model for general chat and light tasks, though it trails most models on everyday questions, drafting and code.
- answering everyday questionsArena Text (overall) · 153rd of 168
- drafting and editing proseArena Creative Writing · 162nd of 168
- writing and completing codeArena Coding · 132nd of 168
EverydayGeneral questions and everyday reasoning
Arena Text (overall)153rd of 168 · 1289
CodingWriting and fixing code on its own
Arena Coding132nd of 168 · 1369
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Not yet scored on Arena Agent.
WritingDrafting and rewriting prose
Arena Creative Writing162nd of 168 · 1204
Arena Creative Writing is the only board that has scored it for this.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done.
Every published score for this model5 scoresEvery figure we hold, from 5 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 15.5 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 23.5 GB spare means a 10% error in the size would not change the answer.
Apple M1 (8-core GPU) · 16 GB
Room to spare. 4.7 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first.
Check against your own machine → · Where to rent it hosted →
Or rent it from someone else
Prices checked 2 hours ago — each listing carries its own date.
The only listing at 131K of context — the other 2 in the table below are not like-for-like. One cheaper row there is outside that comparison: a different quantisation.
- per 1M tokens
- $0.060 in / $0.25 out
- Context served
- 131K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| CoreWeavebf16Through OpenRouter | $0.10 / $0.15checked 2 hours ago | 131K118K max reply | 58 tok/s | No | No | Confirmed |
| OpenRouterOpenRouter's own listing | $0.060 / $0.25checked 2 hours ago | 131K | not measured | Unknown | Unknown | Unknown |
| DeepInfrabf16Direct and through OpenRouter | $0.060 / $0.25checked 2 hours ago | 131K118K max reply through OpenRouter | 25 tok/sthrough OpenRouter | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterNo | DirectUnknownThrough OpenRouterConfirmed |
DeepInfraAcross the 3 listings we hold: 2 say they do not train on prompts (1 of them only through OpenRouter), 0 say they do and 1 does not say. 2 appear in the zero-retention registry we check (1 of them only through OpenRouter); the rest are unknown to us.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| CoreWeavebf16Through OpenRouter | ✓ | ✓ | ✓ |
| OpenRouterOpenRouter's own listing | ✓ | ✓ | ✓ |
| DeepInfrabf16Direct and through OpenRouter | ✓ | ✓ | ✗ |
Tool calling: 3 of 3 listings say yes. JSON output: 3 of 3 listings say yes. Strict schema: 2 of 3 listings say yes, 1 says no.
Models people weigh against Granite 4.2 8B
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 3 listings does not say whether it trains on prompts, and 1 answers only through OpenRouter, not for its own listing.
- We hold no batch or off-peak rate for any of its listings.
- We hold a decode speed for it, but no prompt-processing (prefill) figure, so how long the input side of a job takes is unknown to us.
Licence and identifiers
What the licence allowsApache License 2.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- ibm-granite/granite-4.2-8b
- Architecture
- Dense
- Takes in, gives back
- Text in, text out
- Catalogue slug
- ibm-granite-granite-4-2-8b