IBM Granite Models are a family of enterprise-ready, open foundation models that support multilingual capabilities, coding, retrieval-augmented generation (RAG), tool use, and structured JSON output. Released under Apache 2.0 license.
325K Pulls 48 Tags Updated 2 months ago
Granite Guardian 4.1 is a specialized safety and judging model from IBM Research that evaluates whether LLM prompts and responses meet specified harm criteria.
8,314 Pulls 16 Tags Updated 2 months ago
Granite 4 features improved instruction following (IF) and tool-calling capabilities, making them more effective in enterprise applications.
1.4M Pulls 17 Tags Updated 9 months ago
The IBM Granite 1B and 3B models are long-context mixture of experts (MoE) Granite models from IBM designed for low latency usage.
3M Pulls 33 Tags Updated 1 year ago
IBM Granite 2B and 8B models are 128K context length language models that have been fine-tuned for improved reasoning and instruction-following capabilities.
1.1M Pulls 3 Tags Updated 1 year ago
A compact and efficient vision-language model, specifically designed for visual document understanding, enabling automated content extraction from tables, charts, infographics, plots, diagrams, and more.
969.5K Pulls 5 Tags Updated 1 year ago
The IBM Granite 2B and 8B models are text-only dense LLMs trained on over 12 trillion tokens of data, demonstrated significant improvements over their predecessors in performance and speed in IBM’s initial testing.
1M Pulls 33 Tags Updated 1 year ago
A family of open foundation models by IBM for Code Intelligence
1.5M Pulls 162 Tags Updated 1 year ago
The IBM Granite 2B and 8B models are designed to support tool-based use cases and support for retrieval augmented generation (RAG), streamlining code generation, translation and bug fixing.
The IBM Granite 1B and 3B models are the first mixture of experts (MoE) Granite models from IBM designed for low latency usage.
936.3K Pulls 33 Tags Updated 1 year ago
Granite-3.2 is a family of long-context AI models from IBM Granite fine-tuned for thinking capabilities.
449.2K Pulls 9 Tags Updated 1 year ago
The IBM Granite Embedding 30M and 278M models models are text-only dense biencoder embedding models, with 30M available in English only and 278M serving multilingual use cases.
349.9K Pulls 6 Tags Updated 1 year ago
The IBM Granite Guardian 3.0 2B and 8B models are designed to detect risks in prompts and/or responses.
329.6K Pulls 10 Tags Updated 1 year ago
A general-purpose model ranging from 3 billion parameters to 70 billion, suitable for entry-level hardware.
3M Pulls 119 Tags Updated 2 years ago
Parable is IBM Granite 4.1 fine-tuned on Claude Fable 5 and GPT-5.5 agent traces. Adds think reasoning to Granite. Sibling: Qwen3 line at parable/qwen3-fable
254 Pulls 10 Tags Updated 3 weeks ago
19.6K Pulls 47 Tags Updated 2 months ago
Granite 4.1 models 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.
5,546 Pulls 12 Tags Updated 3 months ago
Granite3.2-Gemma3 is a lightweight, efficient AI model designed for daily conversation, code generation, text analysis and multimodal understanding, delivering stable and responsive performance for various daily tasks.
858 Pulls 4 Tags Updated 4 months ago
169 Pulls 1 Tag Updated 2 months ago
153.4K Pulls 261 Tags Updated 9 months ago