Llama 3.2 Support in Ollama

Meta's Llama 3.2 is now available to run locally via Ollama. This release enables the deployment of lightweight, text-only models for on-device agents and prepares the way for multimodal vision models.

Lightweight Text-Only Models for Edge Devices

Llama 3.2 introduces 1B and 3B parameter models that are specifically optimized for local execution on mobile or edge devices. These models are designed for high-speed, instantaneous responses and the creation of personalized on-device agents.

Key Use Cases for 1B and 3B Models

These text-only models are suited for tasks such as:

  • Summarizing recent communication, such as the last ten messages received on WhatsApp.
  • Summarizing a user's schedule for the coming month.

Because Ollama processes these models locally, user data—including messages and schedules—is not sent to third-party cloud services, ensuring privacy.

Multimodal Vision Capabilities

Ollama will soon support the 11B and 90B Llama 3.2 models. These larger models support image reasoning, which allows them to perform tasks such as:

  • Document-level understanding, including the analysis of charts and graphs.
  • Generating captions for images.

Deployment Instructions

To use Llama 3.2, users must download Ollama and run the following commands:

  • For the 3B model (default): ollama run llama3.2
  • For the 1B model: ollama run llama3.2:1b

Sources

Related

  • Dispatch
    Llama 3.2 Vision available in Ollama

    Ollama has released support for Llama 3.2 Vision in 11B and 90B parameter sizes, enabling local execution of multimodal capabilities including OCR and image analysis.

  • Dispatch
    Llama 3.2 Release Notes: Multimodal Vision and On-Device Small Language Models

    Meta has released Llama 3.2, introducing multimodal vision capabilities in 11B and 90B sizes and lightweight 1B and 3B text-only models optimized for on-device deployment.

  • Dispatch
    Llama 3 Availability on Ollama

    Ollama has released support for Llama 3, providing access to the 8B and 70B parameter models with significant improvements in training data, context length, and token efficiency over Llama 2.

  • Dispatch
    Ollama LLaVA 1.6 Vision Model Support

    Ollama has integrated LLaVA 1.6, enhancing vision capabilities with higher image resolution, improved text recognition, and new model sizes including a 34B parameter version.

  • Dispatch
    Meta Llama 3 Release Notes

    Meta has released Llama 3, an open-access LLM family featuring 8B and 70B parameter models with improved tokenization and training on 15 trillion tokens.