get_usage
Get Usage History
Retrieve API usage history with detailed insights including token counts, response times, costs, source IPs, and model/provider information for each request. Useful for auditing, billing analysis, and optimization.
Ask your agent in plain language. Each of these maps to a real tool on the server.
get_usage
Retrieve API usage history with detailed insights including token counts, response times, costs, source IPs, and model/provider information for each request. Useful for auditing, billing analysis, and optimization.
discover_models
Browse and filter the catalog of available AI models across all providers. Filter by type (language, vision, image, voice, embedding), provider, capabilities, and more. Returns pricing, latency, context window, and availability information for each model.
chat_completion
Send conversational prompts to available AI language models through a unified endpoint. Supports advanced features including web search grounding, RAG augmentation, persistent memory, hallucination reduction via integrity checking, and function calling/tools.
web_search
Perform web searches and scrape web page content. Use search to find relevant web results for a query, and scrape to extract clean text content from specific URLs. These can be used independently or together.
text_to_speech
Convert text into natural-sounding speech using TTS models from OpenAI and ElevenLabs. Access a wide range of voices and customize voice parameters.
generate_image
Create images using AI models from various providers such as DALL-E 3, Stable Diffusion, Flux, and more. Supports generating original images from text prompts and editing existing images.
create_embeddings
Generate vector embeddings from text using embedding models from multiple providers. Useful for semantic search, text clustering, similarity comparison, and retrieval-augmented generation workflows.
manage_rag
Upload documents to RAG collections, list existing collections, and delete collections. Documents are automatically vectorized and can be referenced in chat completions using the "ragTune" parameter to augment model responses with your own data.
get_state
Read saved model, memory, search, routing, and feature settings for the connected application or a specific user.
update_state
Update saved application settings, remove individual setting keys, or control prompt commands and user-scoped state.
analyze_images
Analyze images, extract visible text, compare images, or answer visual questions using a vision-capable AI model.
Every client uses the same Magic MCP URL, and so does any other MCP client. Add it once and sign in with SSO.
{"mcpServers": {"metorial": {"url": "<your-magic-mcp-url>"}}}
claude mcp add --transport http metorial <your-magic-mcp-url>
codex mcp add metorial --url <your-magic-mcp-url>codex mcp login metorial
One governed layer for every app your agents reach, not a separate connector per tool.
Access hundreds of AI models from multiple providers (OpenAI, Anthropic, Google, Meta, etc.) through a unified OpenAI-compatible API. Send chat completions to language models with streaming, function calling, and structured output. Generate images, convert text to speech, analyze images with vision models, and create text embeddings. Augment responses with real-time web search grounding, upload documents for RAG-based retrieval, and reduce hallucinations via integrity checking. Discover and filter available models by type, provider, pricing, and performance. Manage routing preferences (cost or performance optimized), configure model pooling for redundancy, enable persistent conversational memory, and track API usage with detailed cost and token analytics.
Common questions about connecting Apipie Ai to AI agents with Metorial.