This is an AWQ INT4 quantized version of Meta's Llama 3.3 70B Instruct model. It enables efficient local or server-based inference while preserving most of the original model's capabilities, including tool use and reasoning. The model is hosted on Hugging Face.
Meta Llama 3.3 70B Instruct sits in PulseGate's Quantised & converted weights category. It focuses on deploying the large Llama 3.3 70B model with significantly reduced memory footprint via INT4 quantization. It is built as an open-source project for machine learning developers. Meta Llama 3.3 70B Instruct is open source under the MIT license. Meta Llama 3.3 70B Instruct is available on the command line and API.
Behind Meta Llama 3.3 70B Instruct is Meta, based in the United States, and it first shipped in 2023. The GitHub repository has been archived. PulseGate's similarity index places it among 9 comparable projects. Key capabilities include 4-bit quantization, instruct tuned, and tool use support.
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