Overview

predictor.sh exposes OpenAI-compatible and ElevenLabs-compatible APIs, allowing you to use existing SDKs and tools.

Base URL

Your endpoint URL is displayed when you start predictor:

https://abc123.predictor.sh

All API endpoints are relative to this base URL.

Authentication

All requests require a Bearer token in the Authorization header:

curl https://abc123.predictor.sh/v1/chat/completions \
  -H "Authorization: Bearer pred_your_token"

Available Endpoints

Text Generation

Endpoint	Method	Description
/v1/chat/completions	POST	Chat completions
/v1/completions	POST	Text completions
/v1/embeddings	POST	Text embeddings
/v1/models	GET	List models
/health	GET	Health check

Speech-to-Text

Endpoint	Method	Description
/v1/audio/transcriptions	POST	Transcribe audio

Text-to-Speech

Endpoint	Method	Description
/v1/audio/speech	POST	Generate speech (OpenAI)
/v1/text-to-speech/{voice_id}	POST	Generate speech (ElevenLabs)
/v1/text-to-speech/{voice_id}/stream	POST	Streaming speech
/v1/voices	GET	List voices

SDK Compatibility

OpenAI Python SDK

from openai import OpenAI

client = OpenAI(
    base_url="https://abc123.predictor.sh/v1",
    api_key="pred_your_token"
)

# Chat completions
response = client.chat.completions.create(
    model="default",
    messages=[{"role": "user", "content": "Hello!"}]
)

# Audio transcription
transcript = client.audio.transcriptions.create(
    model="whisper-1",
    file=open("audio.mp3", "rb")
)

# Text to speech
speech = client.audio.speech.create(
    model="tts-1",
    voice="alloy",
    input="Hello world"
)

OpenAI JavaScript SDK

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://abc123.predictor.sh/v1',
  apiKey: 'pred_your_token',
});

const response = await client.chat.completions.create({
  model: 'default',
  messages: [{ role: 'user', content: 'Hello!' }],
});

curl

curl https://abc123.predictor.sh/v1/chat/completions \
  -H "Authorization: Bearer pred_your_token" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Error Responses

Errors follow the OpenAI error format:

{
  "error": {
    "message": "Invalid request: missing 'messages' field",
    "type": "invalid_request_error",
    "code": "missing_field"
  }
}

HTTP Status Codes

Code	Description
200	Success
400	Bad request (invalid parameters)
401	Unauthorized (invalid/missing token)
403	Forbidden (IP not allowed)
429	Too many requests
500	Server error
502	Tunnel connection lost
503	Endpoint offline
504	Request timeout