RedHatAI/Qwen2.5-72B-Instruct-FP8-dynamic is a quantized variant of the Qwen2.5 72B instruct model hosted on Hugging Face. It uses dynamic FP8 precision and is provided as part of the open model repository on the platform.
The model includes a specific chat template that defines its instruction-following behavior. The template instructs the model to identify itself as Qwen created by Alibaba Cloud and to act as a helpful assistant. It supports tool calling through an XML-based format where available tools are supplied within tools tags and function calls must be returned inside tool_call tags containing JSON with name and arguments fields. The template handles both cases with and without an initial system message.
It is delivered as a model repository on Hugging Face. The page belongs to the RedHatAI organization and follows the standard structure for models, datasets, and related resources on the site. Hugging Face itself operates to advance and democratize artificial intelligence through open source and open science.
No pricing, licensing details, or additional capabilities are stated on the page.
Qwen2.5 72B Instruct FP8 Dynamic sits in PulseGate's Foundation models & chat category. It focuses on running a 72-billion parameter instruction-tuned model with reduced memory footprint using FP8 quantization. Qwen2.5 72B Instruct FP8 Dynamic is an open-source project aimed at developers. The project is open source (Open Source). It runs on the web and API.
RedHatAI builds and maintains Qwen2.5 72B Instruct FP8 Dynamic, and the product first shipped in 2024. PulseGate's similarity index places it among 5 comparable tools. Among its 3 catalogued features are Large Language Model, Instruction Following, and Quantized Inference.
Latest indexed changes and source events
RedHatAI/Qwen2.5-72B-Instruct-FP8-dynamic verified by the PulseGate indexer
Other apps tracked under the same category.
