# Granite-4.0-H-Small

> Granite-4.0-H-Small is an open-weight language model from IBM, released on 2 Oct 2025. It accepts text and generates text, with a 131K-token context window and up to 131K output tokens per response.

Source page: https://www.aimodel.directory/models/granite-4-h-small
Last verified: 4 Oct 2026

## Specifications

- **Developer:** IBM
- **Release date:** 2 Oct 2025 (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Status:** Generally available (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Weights:** Open weights (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Context window:** 131K (131,072 tokens) (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Max output:** 131K tokens (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Input:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Output:** Text (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Reasoning mode:** No (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Tool calling:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)
- **Structured output:** Yes (source: models.dev, https://github.com/sst/models.dev/blob/dev/providers/watsonx/models/ibm/granite-4-h-small.toml)

## API pricing (USD per 1M tokens)

| Provider | Input | Output | Cached input | Context |
|---|---|---|---|---|

## FAQ

### What is the context window of Granite-4.0-H-Small?

Granite-4.0-H-Small has a context window of 131,072 tokens (131K) and can generate up to 131,072 tokens in a single response, according to models.dev.

### Is Granite-4.0-H-Small open source?

Granite-4.0-H-Small is an open-weight model. Check the license terms before commercial use.

### When was Granite-4.0-H-Small released?

IBM released Granite-4.0-H-Small on 2 Oct 2025, according to models.dev.

---
Data licensed CC BY 4.0. Cite as: AI Model Directory, "Granite-4.0-H-Small", https://www.aimodel.directory/models/granite-4-h-small