Meta

Llama 2 Chat 13B is a 13-billion parameter chat model from Meta's Llama 2 series, fine-tuned with instruction data and Reinforcement Learning from Human Feedback (RLHF). It offers a balance between conversational capability, reasoning performance, and resource efficiency, making it suitable for a wide range of dialogue-based applications.

Llama 2 Chat 70B is Meta's open-source, chat-optimized large language model with 70 billion parameters. It is designed for multi-turn dialogue and excels at complex reasoning and instruction-following tasks.

Llama 2 Chat 7B is an open-source conversational model from Meta, fine-tuned with instruction following and RLHF for interactive applications. As a smaller 7B parameter model, it offers fast inference speeds and low deployment costs while maintaining strong general capabilities.

Meta's latest open-source instruction-tuned model with 70 billion parameters. It supports a 128K context window and demonstrates strong performance in reasoning and coding tasks.

Llama 3 Instruct 8B is a lightweight, instruction-tuned model from Meta's Llama 3 family. It is optimized for fast response times and low-cost deployment, making it suitable for edge devices or applications requiring rapid inference while maintaining strong performance on general tasks.

Llama 3.1 Instruct 405B is Meta's largest and most capable open-source language model, optimized for instruction following and dialogue. It features strong reasoning abilities, supports a 128K token context window, and excels at complex tasks including coding and multilingual understanding.

Llama 3.1 Instruct 70B is a 70-billion parameter, instruction-tuned large language model from Meta. It features a 128K token context window and is optimized for high-performance, open-source deployment across a wide range of tasks.

A lightweight, open-source large language model developed by Meta, supporting a 128K long context window. It is instruction-tuned for dialogue and instruction-following tasks.

This is an 11B parameter multimodal instruct model from Meta's Llama 3.2 series, optimized for vision tasks. It balances performance with efficiency, offering faster inference speeds than larger models, and is suitable for conversational and instruction-following scenarios requiring image understanding.

A lightweight 1B parameter instruction-tuned model from Meta's Llama 3.2 series, optimized for fast inference and deployment on edge devices or resource-constrained environments. It supports text-based tasks and is part of a family that includes larger multimodal models.

A lightweight instruction-tuned model from the Llama 3.2 family, optimized for fast and efficient on-device or edge deployment. It offers low-cost inference with strong multilingual and conversational capabilities for its size.

This is the largest multimodal instruction-tuned model in the Meta Llama 3.2 series, featuring 90 billion parameters and support for both image and text inputs. It excels at visual understanding and complex reasoning tasks, making it suitable for sophisticated applications requiring the processing of both images and text.

Llama 3.3 Instruct 70B is a large, open-source instruction-tuned model from Meta, optimized for complex reasoning, code generation, and following detailed instructions. It features a 128K token context window and delivers strong performance on various benchmarks, positioning it as a powerful and versatile general-purpose assistant.

Llama 4 Maverick is a high-performance, open-weight model from Meta's Llama 4 family, designed for advanced reasoning and coding tasks. It likely features a large context window and multimodal capabilities, continuing the series' focus on powerful, accessible AI.

Llama 4 Scout is a fast and efficient open-source large language model from Meta's Llama series. It is optimized for rapid response times and strong reasoning capabilities, making it suitable for a wide range of general-purpose and coding tasks.

Llama 65B is a large, open-source language model from Meta's Llama family. It provides strong text generation and understanding capabilities, making it a powerful choice for research and enterprise deployment.

Muse Spark is a creative generation model from Meta, designed for rapid ideation and content creation. It excels at producing diverse, high-quality outputs with low latency.