APIClaw

Dify

Use APIClaw inside Dify with the OpenAI-API-compatible model provider plugin.

Prerequisites

  1. Create an API key in the API Keys dashboard.
  2. Copy a full model ID from the model hub.
  3. Make sure Dify can reach https://apiclaw.biz (from Docker, do not use localhost on the host).

Configure

  1. Open your Dify workspace → Integrations → Model Provider (older builds: Settings → Model Providers).
  2. Install / enable the OpenAI-API-compatible plugin if it is not already present.
  3. Add a model and paste this API Base URL:
text
https://apiclaw.biz/v1
  1. Enter:
SettingValue
Model typeLLM
Model nameFull model ID from the model hub
API KeyYour APIClaw API key
API Base URLhttps://apiclaw.biz/v1
Completion modeChat
Model context sizeYour model's real context window
Upper bound for max tokensYour model's real output limit
Function Call Typetool_call
Vision SupportOn, if you send images

Function Call Type defaults to "Not Support", so agents and tool calling fail silently until you set it. The two size fields default to Dify's generic values, which truncate prompts and memory on a model whose real window is larger.

  1. Save and enable the model for your workspace.

For some non-LLM modalities the plugin may append /v1 itself. If embeddings or speech endpoints 404, try the site root for those model types only:

text
https://apiclaw.biz

Verify

Use Dify's model test / preview, then confirm the request in logs.