Dify, a source available platform for building LLM apps, agents, RAG and workflows from a visual console.
Dify is a platform for building and operating applications on top of large language models. Its visual console lets you design AI agents, chat assistants, retrieval augmented generation apps and multi step workflows, connect any model provider, manage prompts and datasets, and publish everything as an API, without stitching together a stack of separate tools. It is distributed under the Dify Open Source License, a modified Apache 2.0 with additional conditions.
It suits teams that want a self hosted workspace to prototype and ship AI features, add retrieval over their own documents, and expose assistants and workflows to their own products through a clean API.
cloudimg ships Dify as its full official container stack, pre wired and hardened behind nginx, always cloning the newest release at build so you start current. There is no default login: the console boots to the first run setup screen so you create your own administrator, and a unique session signing key is generated for your instance on first boot. The stack refreshes to the newest release images on first boot, the paired deployment guide walks through first setup, connecting a model provider and enabling HTTPS, and every instance carries 24/7 support.
Real screenshots taken while testing this image against its deployment guide.
This is a repackaged open source software product wherein additional charges apply for cloudimg support services.
Dify is a popular platform for building and operating LLM applications, distributed under the Dify Open Source License (a modified Apache 2.0 with additional conditions). Its visual console lets you design AI agents, chat assistants, retrieval augmented generation apps and multi step workflows, connect any large language model provider, manage prompts and datasets, and expose everything as an API, without stitching together a stack of separate tools. This image delivers Dify fully installed as its complete official Docker Compose stack, fronted by nginx, so your private AI application workspace is operational within minutes of launch, with no manual container wiring and no dependency troubleshooting.
Unlike a bare deployment, this image ships the whole Dify stack pre-wired and hardened: the console, API, background workers, PostgreSQL, Redis and a Weaviate vector store all on a private Docker network with only nginx exposed on ports 80 and 443, ready for your TLS certificate. There are no default credentials: the image boots to the first-run setup screen so you create your own administrator on first visit, and no shared passwords or pre-seeded database rows exist in the image. A unique session signing key is generated for your instance on first boot. The stack refreshes itself to the newest Dify release images on first boot, so every launch runs current software rather than a version frozen at build time. Docker and systemd manage the services for automatic restarts and clean logging, and 24/7 cloudimg support is included.
Capabilities include a visual workflow and agent builder, retrieval augmented generation over your own documents with a built in vector store, prompt and dataset management, a model provider layer that works with OpenAI, Anthropic, local Ollama and any OpenAI compatible endpoint, built in observability, and a backend as a service API so you can embed AI features into your own products.
cloudimg ships stock Dify with its console logo and copyright retained, and this single tenant image is for customers to self host their own Dify instance, not to operate a multi tenant service. Dify and all other product and company names are trademarks or registered trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them.