Open Systems, Blockchain & Business Daily

OSBBD

AI, money and the software behind them.

Granite 4.1 3B LLMON Preview

Granite 4.1 3B LLMON Preview is an open-weight language model from IBM, released on November 12, 2026. IBM describes it as an experimental preview.

Developer
IBM
Released
November 12, 2026
Total parameters
3B
Active parameters per token
3B
Architecture
Dense
Context window
128K
Input
Text
License
Apache 2.0
Commercial use
Yes
Official weights, GB
7.3 GB
Hugging Face repository
huggingface.co
Checked on

Granite 4.1 3B LLMON Preview is a dense model with 3B parameters, fine-tuned from the Granite 4.1 3B base model. Instead of a conventional chat template it reads prompts written in LLMON (LLM Object Notation), a markup that wraps instructions, data, documents and tool results in separate blocks and states which instruction applies to which input; the bundled chat template converts ordinary messages into this format. The model configuration sets a context window of 131,072 tokens. The model takes text input. The weights are released under the Apache 2.0 license, which allows commercial use.

The official BF16 weights on Hugging Face take about 7.3 GB. Running the model needs at least that much memory across GPUs and system RAM, plus room for the context cache; quantized versions need less.

Total parameters, billions
3B
Active parameters, billions
3B
Context window, tokens
131.07K
License type
Apache 2.0
License conditions
Apache 2.0: keep the license text and notices; includes a patent grant.
Weights precision
BF16
License text
apache.org
Sources
PageURL
Hugging Face model cardhttps://huggingface.co/ibm-granite/granite-4.1-3b-llmon-preview/raw/main/README.md
Hugging Face model confighttps://huggingface.co/ibm-granite/granite-4.1-3b-llmon-preview/raw/main/config.json
Hugging Face file listhttps://huggingface.co/api/models/ibm-granite/granite-4.1-3b-llmon-preview/tree/main
IBM Granite bloghttps://huggingface.co/blog/ibm-granite/llmon-an-llm-native-markup-language-now-in-granite

Questions

What license does Granite 4.1 3B LLMON Preview use?
The weights are released under the Apache 2.0 license, which allows commercial use.
Can I use Granite 4.1 3B LLMON Preview commercially?
Yes. Apache 2.0 allows commercial use if you keep the license and notices.
How much memory does Granite 4.1 3B LLMON Preview need?
The official BF16 weights take about 7.3 GB. Running the model needs at least that much memory across GPUs and system RAM, plus room for the context cache; quantized versions need less. The memory calculator checks which setups fit.
What context window does Granite 4.1 3B LLMON Preview have?
The model configuration sets a context window of 131,072 tokens.

Sources

Every figure comes from the developer’s model card, license text and official announcements, and from the model’s Hugging Face repository metadata. Each model page lists its sources and the date they were checked; the register is rechecked every week.

OSBBD

What are you looking for?

Search by headline, topic or keyword.