Owos/Llama-3.2-1B

Hugging Face
TEXT GENERATIONPricing:Input $0.108 / Output $0.804Concurrent Unit Cost:1Model Size:1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Mar 24, 2026License:llama3.2Architecture:Transformer Featherless Exclusive Warm

The Llama 3.2-1B is a 1.23 billion parameter multilingual large language model developed by Meta, built on an optimized transformer architecture. It is instruction-tuned for multilingual dialogue use cases, excelling in agentic retrieval and summarization tasks. This model is optimized for deployment in constrained environments, such as mobile devices, offering improved inference performance through various quantization schemes.

Loading preview...

Owos/Llama-3.2-1B: Multilingual LLM for Dialogue and On-Device Use

This model is part of Meta's Llama 3.2 collection, a series of multilingual large language models. The Llama 3.2-1B is a 1.23 billion parameter instruction-tuned model, optimized for text-in/text-out generative tasks. It leverages an optimized transformer architecture and has been fine-tuned using supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.

Key Capabilities

  • Multilingual Support: Officially supports English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai, with training on a broader set of languages.
  • Dialogue Optimization: Specifically instruction-tuned for multilingual dialogue use cases, including assistant-like chat.
  • Agentic Applications: Excels in tasks such as knowledge retrieval, summarization, and query/prompt rewriting.
  • Quantization for Efficiency: Features various quantization schemes (SpinQuant, QLoRA) designed for efficient deployment in constrained environments like mobile devices, significantly improving decode speed, time-to-first-token, and reducing model size and memory footprint.
  • Robust Training: Pretrained on up to 9 trillion tokens of publicly available data, with knowledge distillation from larger Llama 3.1 models.

Good For

  • Mobile AI Applications: Its small size and optimized quantization make it suitable for on-device inference with limited computational resources.
  • Multilingual Chatbots and Assistants: Ideal for building conversational AI systems that need to operate across multiple languages.
  • Summarization and Information Retrieval: Effective for tasks requiring the synthesis of information from various sources.
  • Research and Commercial Use: Intended for both research and commercial applications, provided compliance with the Llama 3.2 Community License.

Finetunes
1 models