The page describes a model hosted on Hugging Face with the identifier imagistrali/qwen_2.5_7b-cleanup_cat_ft_numbers_deepJudge_swapped_fullPrompt_seed2. The evidence indicates that it is based on the Qwen architecture, originally created by Alibaba Cloud, and that it is designed to function as a helpful assistant. The chat template references the ability to call functions to assist with user queries, using function signatures provided within XML tags and returning results as JSON objects. The model includes a pad token labeled as <|vision_pad|. No further details about the specific features, intended audience, delivery method, pricing, or licensing are provided in the available evidence. The information does not clarify the exact tasks or domains the model is fine-tuned for, nor does it specify whether it is open source or the terms of its availability. The evidence identifies this tool as a language model implementation available on Hugging Face, with some support for structured function calling within its chat template.
In the Foundation models & chat space, takes a focused approach. It focuses on providing an open-source, fine-tuned language model for advanced text generation and cleanup on local infrastructure. is an open-source project aimed at AI researchers and developers. The project is open source (Apache-2.0). It runs on the command line, and it can be self-hosted.
It is developed by imagistrali, and the product first shipped in 2023. The project is developed in the open on GitHub with 68k stars and 1.1k commits in the last 90 days. Across PulseGate's embedding index, has few near neighbours, marking it as relatively distinct. Among its 5 catalogued features are text generation, text cleanup, and fine-tuned model.
Latest indexed changes and source events
imagistrali/qwen_2.5_7b-cleanup_cat_ft_numbers_deepJudge_swapped_fullPrompt_seed2 discovered by the PulseGate indexer
Other apps tracked under the same category.
