Overview
IBM Granite 4.0 H micro is part of IBMs next-generation line of language models, featuring a hybrid architecture that merges Mamba-2 sequence state modeling with selective transformer layers. The model is a dense variant with 3B total parameters, and enables dramatic improvements in inference and memory efficiency. This model is released under the Apache 2.0 license with cryptographic signing of model checkpoints. Granite 4.0 models are also ISO 42001 certified, making it one of the first open model families to satisfy international standards in trustworthy AI. Granite 4.0 H micro is optimized for long-context and multi-session usage. All Granite 4.0 models have been trained on data samples up to 512K tokens in context length. Performance has been validated on tasks involving context length of up to 128K tokens, but theoretically, the context length can extend further. It is engineered to deliver over 70% reduction in memory usage relative to conventional transformer models in long-context or concurrent workloads. Typical use cases include compact deployments on consumer or enterprise GPUs, edge inference, local experimentation, multi-tool agents, and workloads requiring efficient memory footprint for long input spans. It can also serve as a building block within hybrid systems where its speed and efficiency are exploited for back-end tasks. Granite 4.0 H micro demonstrates competitive performance relative to prior-generation models despite its much smaller active parameter footprint.
Highlights
- Granite 4.0 H micro is fine-tuned using instruction datasets and synthetic data to provide responsive, instruction-following behavior with its efficient architecture. Because the model is architected for long-context and concurrent workloads, it maintains responsiveness and stability even when ingesting large documents or handling multiple sessions in parallel.
- It leverages a hybrid architecture (Mamba-2 + transformer) that omits positional encoding (NoPE), enabling it to generalize to longer contexts without the usual positional bottlenecks found in pure transformer models. Granite 4.0 H micro excels in tasks requiring efficiency and long-context such as summarization of large texts, retrieval-augmented generation (RAG), multi-turn QA, and tool-calling workflows.
- Granite 4.0 H micro supports 12 languages out of the box (English, German, Spanish, French, Japanese, Portuguese, Arabic, Czech, Italian, Korean, Dutch, Chinese), with users able to fine-tune for additional languages.
Details
Unlock automation with AI agent solutions

Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g5.12xlarge Inference (Batch) Recommended | Model inference on the ml.g5.12xlarge instance type, batch mode | $0.00 |
ml.g6e.2xlarge Inference (Real-Time) Recommended | Model inference on the ml.g6e.2xlarge instance type, real-time mode | $0.00 |
ml.g5.24xlarge Inference (Batch) | Model inference on the ml.g5.24xlarge instance type, batch mode | $0.00 |
ml.g6e.4xlarge Inference (Real-Time) | Model inference on the ml.g6e.4xlarge instance type, real-time mode | $0.00 |
ml.g6e.12xlarge Inference (Real-Time) | Model inference on the ml.g6e.12xlarge instance type, real-time mode | $0.00 |
ml.g6e.16xlarge Inference (Real-Time) | Model inference on the ml.g6e.16xlarge instance type, real-time mode | $0.00 |
ml.g6e.24xlarge Inference (Real-Time) | Model inference on the ml.g6e.24xlarge instance type, real-time mode | $0.00 |
ml.g6e.48xlarge Inference (Real-Time) | Model inference on the ml.g6e.48xlarge instance type, real-time mode | $0.00 |
ml.p4d.24xlarge Inference (Real-Time) | Model inference on the ml.p4d.24xlarge instance type, real-time mode | $0.00 |
ml.p5.48xlarge Inference (Real-Time) | Model inference on the ml.p5.48xlarge instance type, real-time mode | $0.00 |
Vendor refund policy
This model is provided by IBM completely free of charge. No payment is required to use it. Therefore, there are no purchases to refund.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Version release notes
The IBM Granite 4.0 H micro model is released under the Apache 2.0 license, offering an architecture combining Mamba-2 and transformer components. It enables striking memory and compute efficiency. The model supports instruction-following, long-context inputs, and multilingual output in 12 languages. It is cryptographically signed to ensure provenance and is part of the Granite 4.0 family which is ISO 42001 certified.
Additional details
Inputs
- Summary
The model can be invoked by passing a prompt. Please see the sample notebook for details.
- Input MIME type
- application/json
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
inputs | The prompt to be passed to the model. | Text | Yes |
Resources
Vendor resources
Support
Vendor support
Support is not provided for this product.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.