www.peerspot.com
Hugging Face reviews 2026 - PeerSpot
Excerpt
|Director/Enterprise Solutions Architect, Technology Advisor at Kyndryl|3.5|I've been using Hugging Face for AI projects and appreciate its versatility and user-friendliness. However, scalability with multi-GPU setups and data cleanup are challenges. I'm also exploring Langchain and Agentic AI to expand my knowledge.| |Student at Renater|4.5|As a student working on personal projects, I find Hugging Face's inference APIs valuable because they save time compared to running inferences locally. However, access to models and datasets could be improved for students and non-professionals.| |Artificial Intelligence Consultant at GlobalLogic|3.5|I primarily use Hugging Face for working with open LLM and embedding models to train and monitor custom data. While its valuable features include rich documentation, it would benefit from a search feature like ChatGPT to assist developers further.| … |Generative AI Developer at Rack Ai Private Limited|4.0|I used Hugging Face to create an SQL chatbot for translating English requests into SQL queries. It's open-source with many packages, but I found the module instructions lacking detail. We resolved code issues using OpenAI embeddings on one project.| |Machine Learning Engineer at TechMinfy|4.0|I use Hugging Face to fine-tune language models for clients due to its ease of use and access to trending open-source models. While improvements are needed in security and documentation, it significantly reduces costs compared to other solutions.|
Related Pain Points
Building secure database access interfaces for non-technical users
7Creating secure admin panels for non-technical users requires juggling encryption, access control, and usability concerns. The complexity rivals building a secondary software system, making it difficult to maintain alongside the primary application.
Scalability challenges with multi-GPU setups
6Enterprise architects report difficulties scaling Hugging Face models across multiple GPUs, limiting the platform's applicability for large-scale production deployments.
Insufficient module documentation and code examples
5Developers report that module instructions lack adequate detail and depth, making it difficult to understand how to properly use specific components without extensive troubleshooting.
Model discovery difficult among millions of models
4With over 2 million models hosted on Hugging Face Hub, finding the right model requires careful manual filtering and semantic search approaches, creating friction in the model selection process.