Loading...
Generate completions, list models, create embeddings, and transcribe audio via OpenAI API
420+ deploysclaude mcp add openai --url https://datafaucet.dev/api/mcp/t/openai
Run this command in your terminal · most users connect in under 60s
POST endpoints
create_chat_completion
Send a prompt to a GPT model. Required: 'model' (gpt-4o, gpt-4o-mini, o1), 'messages' array [{role, content}]. Optional: 'temperature' (0-2), 'max_tokens', 'tools' (function calling), 'response_format' ({type:'json_object'}), 'stream' (true for SSE). Returns choices[].message with content and optional tool_calls.
https://api.openai.com/v1/chat/completions
create_embedding
Generate vector embeddings. Required: 'model' (text-embedding-3-small or text-embedding-3-large), 'input' (string or string[]). Optional: 'dimensions' (256-3072, smaller = faster search). Returns data[].embedding (float array). Batch up to 2048 inputs per call for efficiency.
https://api.openai.com/v1/embeddings
generate_image
Generate images from text. Required: 'prompt' (max 4000 chars for dall-e-3). Optional: 'model' (dall-e-3 or dall-e-2), 'n' (1 for dall-e-3, 1-10 for dall-e-2), 'size' (1024x1024, 1792x1024, 1024x1792), 'quality' (standard or hd), 'response_format' (url or b64_json). Returns data[].url or data[].b64_json.
https://api.openai.com/v1/images/generations
transcribe_audio
Transcribe audio to text. Required: 'file' (multipart upload, max 25MB), 'model' (whisper-1). Optional: 'language' (ISO-639-1), 'response_format' (json, text, srt, verbose_json, vtt), 'timestamp_granularities' (['word','segment']). Supports mp3, mp4, mpeg, mpga, m4a, wav, webm.
https://api.openai.com/v1/audio/transcriptions
create_moderation
Check if text violates OpenAI usage policies. Required: 'input' (string or string[]). Optional: 'model' (omni-moderation-latest). Returns results[] with flagged (bool), categories (hate, violence, sexual, etc.), and category_scores (0-1 floats). Free to call, no token cost.
https://api.openai.com/v1/moderations
GET endpoints
list_models
List all available models. Returns data[] with id, object, created (unix timestamp), and owned_by. Filter client-side by prefix (gpt-4, dall-e, whisper, tts). No pagination needed.
https://api.openai.com/v1/models
Run this in your terminal:
claude mcp add openai https://datafaucet.dev/api/mcp/t/openai
Connect OpenAI API to your AI agent in under 60 seconds. No account required.
{notion_token}) with your real credentialsNeed API keys? Try our free sandbox templates first to see MCP in action with zero setup.
Replace these placeholders with your real values before deploying:
{audio_url}Your value{image_prompt}Your value{openai_api_key}API settings page{prompt}Your value{text}Your valuePreview:
POST https://api.openai.com/v1/chat/completions Authorization: Bearer {openai_api_key} Content-Type: application/json {"model":"gpt-4o","messages":[{"role":"user","content":"{prompt}"}],"max_tokens":1024}
claude mcp add openai --url https://datafaucet.dev/api/mcp/t/openaiWhat your AI agent sees
I have access to 6 tools from OpenAI API: create_chat_completion, list_models, create_embedding, and 3 more.
“Send a prompt to a GPT model. Required: 'model' (gpt-4o, gpt-4o-mini, o1), 'messages' array [{role, content}]. Optional: 'temperature' (0-2), 'max_tokens', 'tools' (function calling), 'response_format' ({type:'json_object'}), 'stream' (true for SSE). Returns choices[].message with content and optional tool_calls.”
Calling create_chat_completion... Done. Here are the results.
Try these prompts
“Generate a completion explaining quantum computing in simple terms”
Copy“List available models and compare their pricing”
Copy“Create an embedding for this paragraph and explain what it represents”
CopyClick any prompt to copy, then paste into your AI client.