Granite 4.1
Granite 4.1
Granite 4.1 models are new and improved granite models which have gone through an improved post-training pipeline, including supervised finetuning and reinforcement learning alignment, resulting in enhanced tool calling, instruction following, and chat capabilities.
Memory Requirements
To run the smallest Granite 4.1, you need at least 2 GB of RAM. The largest one may require up to 18 GB.
Capabilities
Granite 4.1 models support tool use. They are available in gguf.
About Granite 4.1
Granite 4.1 is a family of dense language models offered in 3B, 8B, and 30B parameter sizes. Each model is available in both base and instruction-tuned variants. Compared to Granite 4.0, Granite 4.1 delivers major improvements in tool use, instruction following, coding performance, and mathematical reasoning.
The Granite 4.1 model family
The Granite 4.1 family is available in three sizes tailored for different use cases. The 3B model is compact and optimized for edge deployment and resource-constrained environments. The 8B model offers a balanced mix of performance and efficiency for general-purpose enterprise applications. The 30B model is designed for high-capacity workloads, delivering stronger performance for complex reasoning, advanced coding, and specialized tasks.
License
All Granite 4.1 Language Models are distributed under Apache 2.0 license, allowing free use for both research and commercial purposes.
Docs
See IBM's documentation for running Granite models locally with LM Studio: https://www.ibm.com/granite/docs/run/granite-with-lmstudio.