> For the complete documentation index, see [llms.txt](https://docs.aimlapi.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aimlapi.com/integrations/dify.md).
# Dify
Install the aimlapi.com plugin from the Dify Marketplace and use AI/ML API models across your Dify apps, workflows, and agents.
## About
[Dify](https://dify.ai/) is an open-source platform for building LLM applications — chat apps, agents, and visual workflows — with a built-in RAG pipeline and a model-provider layer that any app in the workspace shares.
AI/ML API ships as an official **Dify Marketplace plugin**, published as [aimlapi/aimlapi](https://marketplace.dify.ai/plugin/aimlapi/aimlapi). Install it from the Marketplace, paste one key, and every app in the workspace can use AI/ML API models.
## Quick start
{% stepper %}
{% step %}
## Open the Marketplace
In Dify, open **Plugins** and switch to the **Marketplace** tab.
{% endstep %}
{% step %}
## Install aimlapi.com
Search for **aimlapi.com** and install the plugin.
{% endstep %}
{% step %}
## Add your API key
Open **Settings → Model Provider**, find **aimlapi.com**, and paste your AI/ML API key.
{% endstep %}
{% step %}
## Pick a model
Select a predefined model, or add a custom model by its exact AI/ML API model ID.
{% endstep %}
{% endstepper %}
## When to use AI/ML API with Dify
AI/ML API works well with Dify when you want:
* one key and one provider entry instead of a separate credential per vendor
* both the chat and embedding models a Dify RAG app needs from a single provider
* to swap the model behind an app without touching the app's prompt or workflow
## Prerequisites
Before you start, make sure you have:
* a Dify workspace — cloud or self-hosted (see [dify.ai](https://dify.ai/))
* an AI/ML API key from [aimlapi.com/app/keys](https://aimlapi.com/app/keys)
* a model ID from [aimlapi.com/models](https://aimlapi.com/models)
Need a key first? Use [API Key Management](broken://pages/cdd447f8b558d5268491de41a1a02c07771cac9b).
## Supported model types
| Dify model type | Used for |
| --------------- | ------------------------------------------- |
| LLM | chat apps, agents, workflow LLM nodes |
| Text Embedding | indexing knowledge into Dify's RAG pipeline |
Image, video, and audio models are not exposed through this plugin.
## Model selection
The plugin ships a generated snapshot of the AI/ML API chat catalog — over 300 predefined models, hottest first — plus embedding models. Anything not in the list can be added as a **custom model**: choose *Add Model*, and enter the full AI/ML API model ID.
### Good starting models
* `openai/gpt-5-5` — strong general use
* `anthropic/claude-sonnet-5` — long-form and agentic work
* `google/gemini-3.6-flash` — lower latency and lower cost
* `openai/text-embedding-3-small` — embeddings
For the full catalog, use [All Model IDs](broken://pages/9db43b2a9ecf64fedb7fcfc447a3ace848669984).
## Config checklist
Make sure these values are set:
* plugin: **aimlapi.com** installed from the Dify Marketplace
* API key: your AI/ML API key, saved in **Settings → Model Provider**
* model: a predefined model, or a custom model with an exact AI/ML API model ID
## Troubleshooting
I cannot find the plugin in the Marketplace
Search for **aimlapi.com**. On a self-hosted Dify, confirm that Marketplace access is enabled for the instance; otherwise install the plugin package manually from the [plugin repository](https://github.com/aimlapi/dify-aimlapi-plugin).
The key saves, but inference fails
A key with a zero balance can still be saved. If the balance is insufficient at inference time, the plugin returns a link to add credits — top up at [aimlapi.com/app](https://aimlapi.com/app).
A model I want is not in the list
The predefined list is a snapshot of the catalog. Add the model as a custom model with its exact AI/ML API model ID, including the vendor prefix.
I get an auth error
Check that:
* the key is valid at [aimlapi.com/app/keys](https://aimlapi.com/app/keys)
* the model ID matches the catalog exactly
* the workspace uses the provider entry you configured, not a leftover custom OpenAI-compatible entry
## Links
* [All Model IDs](broken://pages/9db43b2a9ecf64fedb7fcfc447a3ace848669984)
* [AI/ML API keys](https://aimlapi.com/app/keys)
* [AI/ML API model catalog](https://aimlapi.com/models)
* [aimlapi.com plugin on the Dify Marketplace](https://marketplace.dify.ai/plugin/aimlapi/aimlapi)
* [Dify documentation](https://docs.dify.ai/)
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.aimlapi.com/integrations/dify.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.