n8n community node for Blackbox AI chat models
n8n-nodes-blackbox
bash
npm install n8n-nodes-blackbox
`
Credentials
1. Get your API key from the Blackbox AI Dashboard
2. In n8n, go to Credentials > Add Credential
3. Search for "Blackbox AI API"
4. Enter your API key and save
Supported Models
This node provides access to 100+ AI models through the Blackbox API:
| Provider | Models |
|----------|--------|
| OpenAI | GPT-4o, GPT-5, O3, O4-mini, ChatGPT-4o |
| Anthropic | Claude Sonnet 4.5, Claude Opus 4.5, Claude 3.7 |
| Google | Gemini 3 Pro, Gemini 2.5, Gemma 3 |
| DeepSeek | V3.2, R1, Chat |
| Meta | Llama 4 Maverick/Scout, Llama 3.3 70B |
| Qwen | Qwen3 Max, Qwen3 Coder, QwQ 32B |
| Mistral | Large 2512, Medium 3.1, Codestral, Devstral |
| xAI | Grok 4, Grok 3, Grok Code Fast |
| Cohere | Command A, Command R+ |
| Perplexity | Sonar Pro, Sonar Deep Research |
| Amazon | Nova Premier, Nova Pro |
| And more... | NVIDIA, Minimax, AI21, Moonshot, Z-AI, Arcee, Inflection |
Note on Tool Calling: Not all models support tool/function calling. When using this node with AI Agents that require tools, use models known to support function calling such as GPT-4o, Claude 3.5+, Gemini 2.0+, or Llama 3.1+. Models without tool support will work for basic chat but may fail in agent workflows.
Configuration Options
| Option | Description | Default |
|--------|-------------|---------|
| Temperature | Controls randomness (0-2) | 0.7 |
| Max Tokens | Maximum tokens to generate | 1024 |
| Top P | Nucleus sampling (0-1) | 1 |
| Frequency Penalty | Penalize repeated tokens (-2 to 2) | 0 |
| Presence Penalty | Penalize similar tokens (-2 to 2) | 0 |
Usage
1. Add the Blackbox Chat Model node to your workflow
2. Connect it to an AI Agent or Chain node's Model input
3. Select your preferred model from the dropdown
4. Configure additional options as needed
Technical Details
- Built using @langchain/openai for LangChain compatibility
- Uses Blackbox API endpoint: https://api.blackbox.ai
- Implements INodeType with NodeConnectionTypes.AiLanguageModel` output