Granite 4.1 8B Base
Granite 4.1 8B Base is a pretrained dense decoder model. It provides a starting point for specialized training and supports fill-in-the-middle code completion. It has not received the instruction alignment of the matching chat release.
Model design and training
The model uses grouped-query attention, rotary position embeddings, SwiGLU and shared input/output embeddings. IBM describes roughly 15 trillion tokens of staged pretraining, followed by additional context extension.Granite 4.1 8B Base model card
Practical use
Practical guidance. The 8B base is useful when you need control over the adaptation data and response format. Budget for training and serving separately. A dataset that improves domain terminology can still reduce general instruction following after post-training.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | Downloadable |
| Model identifiers | ibm-granite/granite-4.1-8b-base |
| Context window | 131,072 tokens Published configuration. The card separately describes a 512K context extension. Granite 4.1 8B Base model card |
| License | Apache 2.0Granite 4.1 8B Base model card |
| Parameters | 8BGranite 4.1 8B Base model card |
Related models
- Granite 4.2 8B
IBM · Granite 4.2
An 8B dense text model derived from Granite 4.1 Base with full, low-effort and direct-answer thinking modes.
Launch date Downloadable - Granite 4.1 8B
IBM · Granite 4.1
An instruction-tuned 8B text model for retrieval-based answers, extraction, coding and tool use.
Launch date API · Downloadable · Provider product