Dify
Settings for using Echo Token Station models from the AI app builder Dify.
Dify is an AI app builder in which you can create chatbots and workflows on screen. You add Echo's models with Dify's "OpenAI-API-compatible" plugin (a plugin for services that support the OpenAI format).
Tested: No (as of 2026-09-27). The steps on this page are based on Dify's official documentation and the plugin's description.
What you need
- Your Echo API key (a string that starts with
sk-) - The ID of the model you want to use (for example
deepseek-flash), and that model's context length and maximum output tokens. You can find them in the model list
Set up
- In Dify, open "Integrations" and choose "Model Provider"
- Install the "OpenAI-API-compatible" plugin from the Marketplace
- On the "OpenAI-API-compatible" card, click "Add Model", enter the following and save
| Setting | Value |
|---|---|
| Model Type | LLM |
| Model Name | The ID of the model to use (for example deepseek-flash) |
| Model display name | The name shown in Dify (optional) |
| API Key | Your Echo API key |
| API Base URL | https://token.echohub.io/v1 |
| Completion mode | Chat |
| Model context size | The context length on the model's page (the number of tokens the model can handle at once) |
| Upper bound for max tokens | The maximum output tokens on the model's page |
| Function Call Type | "Tool Call" if you use tool calls (the model calling external functions) |
| Vision Support | "Support" only for models whose page says they accept image input |
The drop-downs and switches in this plugin are labeled in English.
If you run Dify on your own server, that server must be able to connect to https://token.echohub.io.
Check that it works
Choose the model you added in the app settings and talk to it in the preview. If an answer comes back and the call appears in 「呼び出し履歴」 (Call history) in the console, you are done.
Using embedding models
For embeddings (converting text into a sequence of numbers, used for search and knowledge bases), choose "Text Embedding" in "Model Type" and enter the same "API Base URL". You can find models that support embeddings in the model list.
Troubleshooting
| Symptom | Cause | Fix |
|---|---|---|
| 404 with an HTML page instead of JSON when saving | The API Base URL does not end with /v1 | Change it to https://token.echohub.io/v1 |
401 and Invalid API key when saving | The key is wrong, stopped or expired, or has used up its spending limit | Check the key's status in the console |
403 and Insufficient balance | Insufficient balance | Top up |
404 and Invalid URL | The API Base URL has /v1 twice, or goes on to /chat/completions | Change it to https://token.echohub.io/v1 |
503 and The model … does not exist or is not available | The model ID is wrong, or Echo does not offer the model | Check the Model Name |
The list of error bodies is in Errors and fixes.
Sources
| What | Source | Checked |
|---|---|---|
| "Integrations" > "Model Provider", "Add Model" | https://docs.dify.ai/en/use-dify/workspace/model-providers | 2026-09-28 |
| Overview of the OpenAI-API-compatible plugin (version 0.0.68) | https://marketplace.dify.ai/plugin/langgenius/openai_api_compatible | 2026-09-26 |
Setting names and options, appending /chat/completions and /embeddings to the API Base URL | https://github.com/langgenius/dify-official-plugins/tree/main/models/openai_api_compatible | 2026-09-26 |
| HTTP statuses and error bodies on Echo's side | Measured in Echo's test environment (without Dify) | 2026-09-27 |