How to Integrate CCAPI LLM into OpenClaw

This guide walks you through connecting OpenClaw to CCAPI step-by-step. CCAPI is a fully OpenAI-compatible API gateway — OpenClaw treats it as a standard OpenAI endpoint, so no code changes are needed.

OpenClaw is an open-source personal AI assistant that supports multi-platform messaging (WhatsApp, Telegram, Slack, Discord, etc.) and can execute real-world tasks like file management, coding, and web browsing.

Prerequisites

OpenClaw installed — any version that supports custom LLM providers

CCAPI account — sign up at ccapi.ai

CCAPI API Key — create a token in the console

Step 1: Get Your CCAPI Token

Go to ccapi.ai and sign up / log in (Google login supported)

Navigate to Console > Tokens

Click Create Token, name it openclaw

Copy the token key (shown only once)

Step 2: Choose a Model

CCAPI aggregates models from multiple providers at prices well below official rates:

Provider	CCAPI Price	Models
OpenAI	As low as 10% of official	GPT-4o, GPT-5 series, etc.
Google	As low as 20% of official	Gemini 2.5 Flash/Pro, etc.
Anthropic	As low as 20% of official	Claude Sonnet/Opus, etc.
DeepSeek	85% of official	DeepSeek V3.2, etc.
MiniMax	85% of official	MiniMax M2.5, etc.
Zhipu	85% of official	GLM-5, etc.

For the full model list and real-time pricing, visit Model Market.

Recommended models for OpenClaw:

Budget Models (Best Cost-to-Quality)

Model	Provider	Context	Highlights
`gpt-5-nano`	OpenAI	1M	Ultra-low cost, good for simple tasks
`gemini-2.5-flash`	Google	1M	Fast, large context window
`gpt-4o-mini`	OpenAI	128K	Balanced choice
`deepseek-v3.2`	DeepSeek	128K	Strong coding and reasoning
`minimax-m2.5`	MiniMax	1M	1M context, excellent at coding

Mid-Tier Models (Quality + Affordability)

Model	Provider	Context	Highlights
`gpt-5-mini`	OpenAI	1M	GPT-5 family, fast and reliable
`glm-5`	Zhipu	32K	Excellent Chinese language capabilities
`gemini-2.5-pro`	Google	1M	Google's strongest model, multimodal

Premium Models (Maximum Quality)

Model	Provider	Context	Highlights
`gpt-4o`	OpenAI	128K	Multimodal, proven quality
`claude-sonnet-4-5`	Anthropic	200K	Anthropic flagship
`gpt-4.1`	OpenAI	1M	1M context window

Vision-Capable Models (for screenshot analysis)

If you use OpenClaw's screenshot/vision features, choose a vision-capable model:

Model	Provider	Highlights
`gemini-2.5-flash`	Google	Cheapest vision option
`gpt-5-mini`	OpenAI	Best vision value
`gpt-4o`	OpenAI	Best overall vision quality
`gemini-2.5-pro`	Google	Google's best vision

Step 3: Configure OpenClaw

There are three ways to connect OpenClaw to CCAPI. Choose the one that fits your workflow.

Method A: Environment Variables (Simplest)

Add these to your shell profile (~/.bashrc, ~/.zshrc) or OpenClaw's .env file:

Variable	Required	Description
`OPENCLAW_BASE_URL`	Yes	Must be `https://api.ccapi.ai/v1`
`OPENCLAW_API_KEY`	Yes	Your CCAPI token
`OPENCLAW_MODEL`	Yes	Default model, e.g. `deepseek-v3.2`
`OPENCLAW_FALLBACK_MODEL`	No	Backup model, auto-switches on failure

Restart your terminal and OpenClaw after setting the variables.

Method B: Config File (Most Flexible)

Edit ~/.openclaw/openclaw.json for multi-model setups:

{
  models: {
    mode: "merge",
    providers: {
      ccapi: {
        baseUrl: "https://api.ccapi.ai/v1",
        apiKey: "${CCAPI_API_KEY}",
        api: "openai-completions",
        models: [
          {
            id: "deepseek-v3.2",
            name: "DeepSeek V3.2",
            reasoning: false,
            input: ["text"],
            contextWindow: 128000,
            maxTokens: 32000,
          },
          {
            id: "gpt-4o",
            name: "GPT-4o",
            reasoning: false,
            input: ["text", "image"],
            contextWindow: 128000,
            maxTokens: 16384,
          },
          {
            id: "gpt-5-mini",
            name: "GPT-5 Mini",
            reasoning: false,
            input: ["text", "image"],
            contextWindow: 1000000,
            maxTokens: 32000,
          },
          {
            id: "gemini-2.5-flash",
            name: "Gemini 2.5 Flash",
            reasoning: false,
            input: ["text", "image"],
            contextWindow: 1000000,
            maxTokens: 65536,
          },
          {
            id: "minimax-m2.5",
            name: "MiniMax M2.5",
            reasoning: false,
            input: ["text"],
            contextWindow: 1000000,
            maxTokens: 32000,
          },
          {
            id: "glm-5",
            name: "GLM-5",
            reasoning: false,
            input: ["text"],
            contextWindow: 32000,
            maxTokens: 4096,
          },
          {
            id: "claude-sonnet-4-5",
            name: "Claude Sonnet 4.5",
            reasoning: true,
            input: ["text", "image"],
            contextWindow: 200000,
            maxTokens: 64000,
          },
        ],
      },
    },
  },
  agents: {
    defaults: {
      model: {
        primary: "ccapi/deepseek-v3.2",
      },
    },
  },
}

Field	Description
`models.providers.ccapi`	Registers CCAPI as a custom provider named "ccapi"
`baseUrl`	CCAPI endpoint, always `https://api.ccapi.ai/v1`
`apiKey`	Use `"${CCAPI_API_KEY}"` to reference an env var (more secure), or paste the token directly
`api`	Must be `"openai-completions"` — tells OpenClaw to use the OpenAI protocol
`models`	Model catalog for this provider. Only listed models appear in OpenClaw's model selector
`agents.defaults.model.primary`	Default model. Format: `provider-name/model-id`, e.g. `ccapi/deepseek-v3.2`

Tip: mode: "merge" combines with built-in catalogs; mode: "replace" fully overwrites them.

Method C: CLI Commands (Quick Setup)

Verify the config:

Step 4: Verify the Integration

4.1 Test the API Key Directly

You should get a JSON response with choices[0].message.content.

4.2 Test Streaming

4.3 Test in OpenClaw

Launch OpenClaw and give it a simple task:

What is 2 + 2?

If OpenClaw responds normally, the integration is working. You can check the call logs at CCAPI Console.

4.4 Test Tool Calling

OpenClaw relies heavily on tool/function calling (shell commands, file operations, etc.). Verify:

List the files in the current directory.

OpenClaw should call a shell tool to run ls and return the output. All CCAPI text models support function calling.

Step 5: Optimize Your Setup

Recommended Configurations

Daily automation (~$3/month)

Best for: file management, web browsing, shell scripts, everyday tasks.

Coding assistant (~$8/month)

Best for: code review, debugging, refactoring.

Quality-first (~$15/month)

Best for: complex reasoning, multi-step tasks, screenshot analysis.

Chinese language tasks

Best for: Chinese documentation, code with Chinese comments, Chinese web browsing.

Cost Comparison

Using the same model usage, calling through CCAPI significantly reduces costs:

Provider	Direct Official	Via CCAPI	Savings
OpenAI	Official price	As low as 10%	Save up to 90%+
Google Gemini	Official price	As low as 20%	Save up to 78%
Anthropic Claude	Official price	As low as 20%	Save up to 78%
DeepSeek	Official price	85%	Save 15%

CCAPI is pay-as-you-go with no subscription fees. For specific pricing, visit Model Market.

Troubleshooting

Problem	Cause	Solution
`Invalid token`	Token is incorrect or expired	Check token status at Console > Tokens
`Token quota exhausted`	Account balance is $0	Top up at Console > Top Up
`Model not found`	Model name typo	Check model name spelling. View available models at Model Market
OpenClaw still uses OpenAI	`OPENCLAW_BASE_URL` not set	Verify with `echo $OPENCLAW_BASE_URL` — must output `https://api.ccapi.ai/v1`
Tool calls fail	Model doesn't support function calling	All CCAPI text models support function calling. Check the model name is correct
Streaming doesn't work	Network proxy or firewall blocking SSE	Test with `curl -N`. Check proxy settings
Slow responses	Model overloaded or network latency	Try `gemini-2.5-flash` (fastest model)

Quick Reference

Base URL: https://api.ccapi.ai/v1

Auth: Authorization: Bearer your-ccapi-token

Endpoint: POST /v1/chat/completions

Supported features: streaming, tool/function calling, vision (image input), JSON mode, system messages

Full Documentation: docs.ccapi.ai

Model Market: ccapi.ai/pricing

Console: ccapi.ai/console

Top Up: ccapi.ai/console/topup

Logs: ccapi.ai/console/log

OpenClaw + CCAPI LLM Integration Guide

How to Integrate CCAPI LLM into OpenClaw#

Prerequisites#

Step 1: Get Your CCAPI Token#

Step 2: Choose a Model#

Budget Models (Best Cost-to-Quality)#

Mid-Tier Models (Quality + Affordability)#

Premium Models (Maximum Quality)#

Vision-Capable Models (for screenshot analysis)#

Step 3: Configure OpenClaw#

Method A: Environment Variables (Simplest)#

Method B: Config File (Most Flexible)#

Method C: CLI Commands (Quick Setup)#

Step 4: Verify the Integration#

4.1 Test the API Key Directly#

4.2 Test Streaming#

4.3 Test in OpenClaw#

4.4 Test Tool Calling#

Step 5: Optimize Your Setup#

Recommended Configurations#

Cost Comparison#

Troubleshooting#

Quick Reference#