Set up Dify
Build AI apps and workflows; install the OpenAI-compatible plugin to use CherryChat models
- API format: OpenAI Chat Completions
- Base URL format: Up to /v1 (https://gpt.cherrychat.org/v1)
- Recommended model: gpt-6.1-sol, deepseek-v4-flash
- Platforms: Web
Steps
Install the OpenAI-API-compatible plugin
Go to Integrations → Model Provider (older versions: avatar at top right → Settings → Model Provider), find "OpenAI-API-compatible" under Install model providers and install it.
Add a model
On the OpenAI-API-compatible card click Add Model, set the model type to LLM and fill in the fields below.
- Model Name: gpt-6.1-sol (The model ID, not a display name)
- API Key: sk-your-key
- API Base URL: https://gpt.cherrychat.org/v1 (End at /v1 with no trailing /)
Set the capability options
Completion mode: Chat. Model context size defaults to only 4096; raise it per the Models page (e.g. 128000). Upper bound for max tokens: about 8192. Function Call Type: Tool Call, so agents and workflows can use tools. Vision Support only for vision models.
Save and set as default
Save; Dify validates the key and URL first. Then open Default Model at the top right and pick it as the system reasoning model, or choose it directly in an app or workflow LLM node.
Embedding for knowledge bases (optional)
For knowledge bases, click Add Model again with type Text Embedding, model name text-embedding-3-small, and the same API key and Base URL.
Common pitfalls
- API Base URL is https://gpt.cherrychat.org/v1: no trailing slash and no /chat/completions.
- Model Name is the model ID (lowercase, as on the Models page), not a display name; use "Model display name" for that.
- Model context size defaults to 4096; leave it and long documents or conversations get truncated.
- For self-hosted Dify, make sure the server can reach https://gpt.cherrychat.org; Dify in Docker uses the server's own network.
Error → cause → fix
| What you see | Why | What to do |
|---|---|---|
| Invalid credentials / 401 on save | The key is wrong, has spaces, or was disabled | Copy the key again from My Keys and save again |
| 404 on save | Base URL is missing /v1, has a trailing slash, or is https://gpt.cherrychat.org/v1/chat/completions | Change it to https://gpt.cherrychat.org/v1 |
| Model not found | Model Name holds a display name or a typo | Use the exact model ID from the Models page, e.g. gpt-6.1-sol |
| Self-hosted: connection keeps timing out | The Dify server cannot reach https://gpt.cherrychat.org | Run curl https://gpt.cherrychat.org/v1/models on the server; any JSON reply, even an error, means the network path works |
| HTTP 402 or "insufficient balance" (billing_error in Claude format) | Your credits cannot cover this call. API calls are paid from credits only; memberships and unlimited plans do not apply | Top up any amount in the CherryChat Store, then retry after a few seconds |
| HTTP 429 Too Many Requests | Too many requests: by default 120 per minute per key and 8 at a time per account | Lower the concurrency or requests per second in the tool and wait for the Retry-After seconds; most tools retry on their own |
Guide checked 2026-11-02. The tool's screens may change; the official docs win if they differ.
Operated by UK Company CHERRYCHAT LTD · Company Number 16096119 · Support WeChat: JsnonoChat · Content is AI-generated — please verify before relying on it
