Granite 4.2 8B
Granite 4.2 8B is a large language model made by IBM, added to the catalog on 31 Aug 2026. It has a context window of 131,072 tokens and costs $0.06 per 1M input tokens and $0.15 per 1M output tokens. It accepts text and produces text. Typical uses: mathematical reasoning, code generation and multilingual dialogue.
Price checked ยท source: OpenRouter
What Granite 4.2 8B is
IBM Granite 4.2 8B is a dense reasoning model from IBM. It accepts text and produces text. The provider describes it as suited to mathematics, code generation, multilingual dialogue, and agentic workflows that need multi-step reasoning. Its listed modalities are text input and text output; the supplied facts do not list image or audio input or output. The model supports reasoning, tools, structured outputs, and response_format. The catalog does not specify which tools or output schemas are available, so these capabilities indicate support for workflows that use tools or request structured responses, not a particular integration.
It has a context length of 131,072 tokens and a maximum output of 117,964 tokens. Context length describes how much input the model can handle in a request, while the output limit describes the maximum size of a generated response. These limits suit requests involving substantial text or tasks where a longer response is needed, provided each stays within its respective limit. Its stated areas of use include mathematics, code generation, multilingual text dialogue, and multi-step workflows. The facts do not name supported languages, programming languages, specific tools, or guarantees for particular tasks. It is a text-only option for work that calls for reasoning, tool support, or structured outputs, with a large stated context and output allowance.
Key features
- Accepts text and produces text
- Has a 131,072-token context length
- Supports a maximum output of 117,964 tokens
- Supports tools, structured outputs, and response_format
- Supports reasoning
Typical uses
- Mathematical reasoning
- Code generation
- Multilingual dialogue
- Multi-step agentic workflows
Granite 4.2 8B pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.06 |
| Output, per 1M tokens | $0.15 |
| Cached input read, per 1M tokens | $0.015 |
No price change since it was added on 31 Aug 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0001 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0016 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0021 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.0092 |
Granite 4.2 8B specifications
| Model id | ibm-granite/granite-4.2-8b |
|---|---|
| Version | granite-4.2-8b-20260831 |
| Context window | 131,072 tokensAbout 98,304 English words |
| Max output | 117,964 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 31 Aug 2026 |
| Input types | text |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | Yes |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Granite 4.2 8B yet.