Downloadmeta-llama/Llama-3.2-1B is a lightweight, instruction-tuned generative language model developed by Meta, optimized for multilingual dialogue, summarization, and retrieval tasks. With 1.23 billion parameters, it offers strong performance in constrained environments like mobile devices, without sacrificing versatility or multilingual support. It is part of the Llama 3.2 family, trained on up to 9 trillion tokens and aligned using supervised fine-tuning, preference optimization, and safety tuning. The model supports eight officially listed languages (including Spanish, German, Hindi, and Thai) but can be adapted to more. Llama 3.2-1B outperforms other open models in several benchmarks relative to its size and offers quantized versions for efficiency. It uses a refined transformer architecture with Grouped-Query Attention (GQA) and supports long context windows of up to 128k tokens.
Features
- Pretrained and instruction-tuned for assistant-like applications
- Supports 8+ languages, with multilingual inputs and outputs
- 1.23B parameters optimized for low-resource environments
- Long context support (up to 128k tokens)
- Quantized variants for mobile and on-device inference
- Aligned using SFT, DPO, and safety fine-tuning
- Competitive scores on industry benchmarks like MMLU, ARC, and TLDR
- Includes system-level tools like Llama Guard and Prompt Guard
Project Samples
