Create a scoped key
Generate a cm_ API key and grant only the service permissions the application needs.
Use one CallMissed API key for model inference, speech, images, and search, then connect those capabilities to voice and WhatsApp products.
Each capability has a documented endpoint, request schema, model catalog, and playground path.
POST /v1/chat/completionsUse OpenAI-compatible chat requests with streaming, tools, and model-specific capabilities.
Read endpoint guidePOST /v1/audio/transcriptionsUpload audio for transcription, language-aware speech recognition, or translation workflows.
Read endpoint guidePOST /v1/audio/speechGenerate speech from text with the voices, languages, and formats supported by the selected model.
Read endpoint guidePOST /v1/images/generationsCreate images through an OpenAI-compatible request shape and a catalog of image models.
Read endpoint guidePOST /v1/searchRun current web searches through a dedicated endpoint with source-aware results.
Read endpoint guideGenerate a cm_ API key and grant only the service permissions the application needs.
Pick the API surface and model that match the input, output, latency, and price requirements.
Try the request in the playground, then review usage and errors from the dashboard.
curl https://api.callmissed.com/v1/chat/completions \
-H "Authorization: Bearer cm_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "sarvam-105b",
"messages": [{"role": "user", "content": "Namaste! Order status?"}]
}'Authentication uses Authorization: Bearer cm_YOUR_KEY. Keep keys server-side and use separate scoped keys for separate applications.
Create an API key and send your first authenticated request.
Full API reference for LLM, STT, TTS, voice, WhatsApp, and webhooks.
Test models, STT, TTS, and voice agents without writing code.
OpenAI-compatible — use official OpenAI SDKs with base URL api.callmissed.com.
New models, endpoints, and breaking changes.
Need help? Help Center · Contact sales