Llama-4-Maverick-17B-128E-Instruct is a 17-billion parameter (with 128 experts) instruction-tuned model from Meta's Llama 4 series. It supports advanced capabilities including tool calling, long context, and high-quality dialogue. The model weights are hosted on Hugging Face for local deployment and fine-tuning.
In the Text generation space, Llama 4 Maverick 17B 128E Instruct takes a focused approach. It focuses on delivering a high-performance open mixture-of-experts model for complex reasoning and instruction tasks. It is built as an open-source project for developers. The project is open source (Open Source). It ships for the web and the command line, and it can be self-hosted.
It is developed by Meta (United States), and it first shipped in 2024. Development happens publicly on GitHub with 7.7k stars. Key capabilities include mixture of Experts, Instruction Following, and Tool Use.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do