Llama-3.2-3B

Llama 3.2 3B\

The Llama 3.2 3B model is a lightweight, text-only version of the Llama 3.2 model, designed to be highly capable with multilingual text generation and tool-calling abilities.

Key Features

• Highly capable with multilingual text generation\
• Tool-calling abilities for direct interaction with external tools and services\
• Optimized for edge and mobile devices\
• Supports context length of 128K tokens\
• Available for fine-tuning and deployment on a variety of platforms\
• Part of the Llama 3.2 ecosystem, providing seamless integration with other Llama models

Technical Specifications

• Model size: 3B parameters\
• Context length: 128K tokens\
• Input type: Text\
• Output type: Text\
• Pre-trained on: Large-scale noisy (text) pair data\
• Fine-tuned on: Medium-scale high-quality in-domain and knowledge-enhanced (text) pair data\
• Weights: Based on BFloat16 numerics

Performance Metrics

• Outperforms Gemma 2 2.6B and Phi 3.5-mini models on tasks such as following instructions, summarization, prompt rewriting, and tool-use\
• Competitive with Gemma 2 2.6B model on tasks such as summarization and prompt rewriting

Use Cases

• Personalized on-device agentic applications with strong privacy\
• Text summarization and generation\
• Instruction following and prompt rewriting\
• Tool-calling for direct interaction with external tools and services\
• Multilingual text generation and translation

Serve with vLLM

vLLM is a fast and easy-to-use library for LLM inference and serving.

vLLM is fast with:

• State-of-the-art serving throughput
• Efficient management of attention key and value memory with PagedAttention
• Continuous batching of incoming requests
• Fast model execution with CUDA/HIP graph
• Quantization: GPTQ, AWQ, SqueezeLLM, FP8 KV Cache
• Optimized CUDA kernels

vLLM is flexible and easy to use with:

• Seamless integration with popular Hugging Face models
• High-throughput serving with various decoding algorithms, including parallel sampling, beam search, and more
• Tensor parallelism support for distributed inference
• Streaming outputs
• OpenAI-compatible API server
• Support NVIDIA GPUs and AMD GPUs
• (Experimental) Prefix caching support
• (Experimental) Multi-lora support

vLLM seamlessly supports many Hugging Face models, including the following architectures:

• Aquila & Aquila2 (BAAI/AquilaChat2-7B, BAAI/AquilaChat2-34B, BAAI/Aquila-7B, BAAI/AquilaChat-7B, etc.)
• Baichuan & Baichuan2 (baichuan-inc/Baichuan2-13B-Chat, baichuan-inc/Baichuan-7B, etc.)
• BLOOM (bigscience/bloom, bigscience/bloomz, etc.)
• ChatGLM (THUDM/chatglm2-6b, THUDM/chatglm3-6b, etc.)
• Command-R (CohereForAI/c4ai-command-r-v01, etc.)
• DBRX (databricks/dbrx-base, databricks/dbrx-instruct etc.)
• DeciLM (Deci/DeciLM-7B, Deci/DeciLM-7B-instruct, etc.)
• Falcon (tiiuae/falcon-7b, tiiuae/falcon-40b, tiiuae/falcon-rw-7b, etc.)
• Gemma (google/gemma-2b, google/gemma-7b, etc.)
• GPT-2 (gpt2, gpt2-xl, etc.)
• GPT BigCode (bigcode/starcoder, bigcode/gpt_bigcode-santacoder, etc.)
• GPT-J (EleutherAI/gpt-j-6b, nomic-ai/gpt4all-j, etc.)
• GPT-NeoX (EleutherAI/gpt-neox-20b, databricks/dolly-v2-12b, stabilityai/stablelm-tuned-alpha-7b, etc.)
• InternLM (internlm/internlm-7b, internlm/internlm-chat-7b, etc.)
• InternLM2 (internlm/internlm2-7b, internlm/internlm2-chat-7b, etc.)
• Jais (core42/jais-13b, core42/jais-13b-chat, core42/jais-30b-v3, core42/jais-30b-chat-v3, etc.)
• LLaMA, Llama 2, and Meta Llama 3 (meta-llama/Meta-Llama-3-8B-Instruct, meta-llama/Meta-Llama-3-70B-Instruct, meta-llama/Llama-2-70b-hf, lmsys/vicuna-13b-v1.3, young-geng/koala, openlm-research/open_llama_13b, etc.)
• MiniCPM (openbmb/MiniCPM-2B-sft-bf16, openbmb/MiniCPM-2B-dpo-bf16, etc.)
• Mistral (mistralai/Mistral-7B-v0.1, mistralai/Mistral-7B-Instruct-v0.1, etc.)
• Mixtral (mistralai/Mixtral-8x7B-v0.1, mistralai/Mixtral-8x7B-Instruct-v0.1, mistral-community/Mixtral-8x22B-v0.1, etc.)
• MPT (mosaicml/mpt-7b, mosaicml/mpt-30b, etc.)
• OLMo (allenai/OLMo-1B-hf, allenai/OLMo-7B-hf, etc.)
• OPT (facebook/opt-66b, facebook/opt-iml-max-30b, etc.)
• Orion (OrionStarAI/Orion-14B-Base, OrionStarAI/Orion-14B-Chat, etc.)
• Phi (microsoft/phi-1_5, microsoft/phi-2, etc.)
• Phi-3 (microsoft/Phi-3-mini-4k-instruct, microsoft/Phi-3-mini-128k-instruct, etc.)
• Qwen (Qwen/Qwen-7B, Qwen/Qwen-7B-Chat, etc.)
• Qwen2 (Qwen/Qwen1.5-7B, Qwen/Qwen1.5-7B-Chat, etc.)
• Qwen2MoE (Qwen/Qwen1.5-MoE-A2.7B, Qwen/Qwen1.5-MoE-A2.7B-Chat, etc.)
• StableLM(stabilityai/stablelm-3b-4e1t, stabilityai/stablelm-base-alpha-7b-v2, etc.)
• Starcoder2(bigcode/starcoder2-3b, bigcode/starcoder2-7b, bigcode/starcoder2-15b, etc.)
• Xverse (xverse/XVERSE-7B-Chat, xverse/XVERSE-13B-Chat, xverse/XVERSE-65B-Chat, etc.)
• Yi (01-ai/Yi-6B, 01-ai/Yi-34B, etc.)

Getting Started

Visit our documentation to get started.

• Installation
• Quickstart
• Supported Models