Open Mind API
๐Ÿš€ 480+ Models Active & Free ยท Inception Labs Mercury-3 1M Diffusion Added

Self-Hosted Open Mind API

Access frontier and open-source models from LM Arena, Inception Labs (Mercury-3 Diffusion 1M), Groq, Cloudflare, OpenRouter, Nvidia NIM, BluesMinds, AgentRouter, Google Gemini, Ollama, and OmniRoute Gateway via a unified, zero-latency, OpenAI-compatible endpoint. Completely free.

Authentication

Pass a Bearer token in your headers to authenticate requests. The base URL acts as a drop-in replacement for OpenAI.

๐ŸŒ

Base URL

https://openmind.theeverythingai.com/v1

Change your client base URL config to route queries through our server.

๐Ÿ”‘

Bearer Token

Authorization: Bearer YOUR_API_KEY

Pass your self-service API key in the authorization headers.

Keys & Usage Dashboard

Register a new API key instantly or check current usage and rate limits.

๐Ÿ” Generate Free API Key

Instant setup. One key per email.
โš ๏ธ Click key to copy. Copy now โ€” it will not be shown again.

๐Ÿ“Š API Usage Tracker

Monitor real-time token statistics.
POST /v1/chat/completions

Chat Completions

Generate completions for a list of messages. Supports text, multimodal inputs, streaming responses, and reasoning models.

Parameters

FieldTypeDescription
model string Required. The ID of the model to use (e.g. llama-3.3-70b-versatile).
messages array Required. An array of message objects: {role, content}.
temperature number Optional. Sampling temperature (0.0 to 2.0). Defaults to 0.7.
max_tokens integer Optional. Maximum tokens to generate.
stream boolean Optional. Stream response chunks via server-sent events. Defaults to false.
curl https://openmind.theeverythingai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "llama-3.3-70b-versatile",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://openmind.theeverythingai.com/v1"
)

response = client.chat.completions.create(
    model="llama-3.3-70b-versatile",
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)
print(response.choices[0].message.content)
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://openmind.theeverythingai.com/v1',
});

const response = await client.chat.completions.create({
  model: 'llama-3.3-70b-versatile',
  messages: [
    { role: 'user', content: 'Hello!' }
  ],
});
console.log(response.choices[0].message.content);
# Inception Labs Mercury-3: 1M-context continuous diffusion LLM with deep reasoning
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://openmind.theeverythingai.com/v1"
)

response = client.chat.completions.create(
    model="mercury-3",
    messages=[
        {"role": "user", "content": "What is a diffusion model and how does continuous diffusion work?"}
    ]
)
print(response.choices[0].message.content)
POST /v1/embeddings

Embeddings

Generate vector embeddings for input text. Ideal for RAG, semantic search, and clustering tasks.

Parameters

FieldTypeDescription
model string Required. The ID of the embedding model (e.g. nomic-embed-text:latest).
input string or array Required. The input text to embed, or an array of strings.
curl https://openmind.theeverythingai.com/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "nomic-embed-text:latest",
    "input": "Vector search is powerful!"
  }'
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://openmind.theeverythingai.com/v1"
)

response = client.embeddings.create(
    model="nomic-embed-text:latest",
    input="Vector search is powerful!"
)
print(response.data[0].embedding)
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://openmind.theeverythingai.com/v1',
});

const response = await client.embeddings.create({
  model: 'nomic-embed-text:latest',
  input: 'Vector search is powerful!',
});
console.log(response.data[0].embedding);
POST /v1/rerank

Rerank Documents

State-of-the-art multilingual neural cross-encoder powered by Jina AI. Re-ranks candidate documents by exact semantic relevance to a query for high-precision RAG and search pipelines.

Parameters

FieldTypeDescription
model string Optional. Reranker model ID. Default: jina-reranker-v2-base-multilingual.
query string Required. The search query or question.
documents array Required. Array of text strings or document objects to score and rank.
top_n integer Optional. Number of top ranked results to return (e.g. 3).
return_documents boolean Optional. Whether to include document text in results (default: true).

Response Format

Returns a ranked list of scored documents with normalized relevance scores:

{
  "model": "jina-reranker-v2-base-multilingual",
  "results": [
    {
      "index": 1,
      "relevance_score": 0.8081,
      "document": { "text": "The cheetah is the fastest land animal..." }
    }
  ]
}
curl https://openmind.theeverythingai.com/v1/rerank \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "jina-reranker-v2-base-multilingual",
    "query": "Which animal is the fastest on land?",
    "documents": [
      "The peregrine falcon is the fastest bird diving at 300 km/h.",
      "The cheetah is the fastest land animal, reaching 120 km/h.",
      "The sailfish is the fastest animal in the ocean."
    ],
    "top_n": 2,
    "return_documents": true
  }'
import requests

url = "https://openmind.theeverythingai.com/v1/rerank"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}
payload = {
    "model": "jina-reranker-v2-base-multilingual",
    "query": "Which animal is the fastest on land?",
    "documents": [
        "The peregrine falcon is the fastest bird diving at 300 km/h.",
        "The cheetah is the fastest land animal, reaching 120 km/h.",
        "The sailfish is the fastest animal in the ocean."
    ],
    "top_n": 2,
    "return_documents": True
}

resp = requests.post(url, headers=headers, json=payload)
data = resp.json()
for r in data["results"]:
    print(f"Rank score: {r['relevance_score']:.4f} -> {r['document']}")
const response = await fetch("https://openmind.theeverythingai.com/v1/rerank", {
  method: "POST",
  headers: {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    model: "jina-reranker-v2-base-multilingual",
    query: "Which animal is the fastest on land?",
    documents: [
      "The peregrine falcon is the fastest bird diving at 300 km/h.",
      "The cheetah is the fastest land animal, reaching 120 km/h.",
      "The sailfish is the fastest animal in the ocean."
    ],
    top_n: 2,
    return_documents: true
  })
});
const data = await response.json();
console.log(data.results);
POST /v1/audio/speech

Audio Speech (TTS)

Convert text to high-quality spoken audio. Features multiple voices and models supported via Cloudflare Worker AI and Groq TTS.

Parameters

FieldTypeDescription
model string Required. The ID of the TTS model (e.g. @cf/myshell-ai/melotts or canopylabs/orpheus-v1-english).
input string Required. The text content to generate speech for. Max 4096 characters.
voice string Optional. The voice ID (e.g., troy or fahad for Groq models). Defaults to model preference.
curl https://openmind.theeverythingai.com/v1/audio/speech \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -o speech.wav \
  -d '{
    "model": "@cf/myshell-ai/melotts",
    "input": "Welcome to Open Mind API. Speech generation is live!"
  }'
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://openmind.theeverythingai.com/v1"
)

response = client.audio.speech.create(
    model="@cf/myshell-ai/melotts",
    voice="troy",
    input="Welcome to Open Mind API. Speech generation is live!"
)
response.stream_to_file("speech.wav")
import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://openmind.theeverythingai.com/v1',
});

const mp3 = await client.audio.speech.create({
  model: '@cf/myshell-ai/melotts',
  voice: 'troy',
  input: 'Welcome to Open Mind API. Speech generation is live!',
});

const buffer = Buffer.from(await mp3.arrayBuffer());
await fs.promises.writeFile('speech.wav', buffer);
POST /v1/audio/transcriptions

Audio Transcriptions

Transcribe spoken audio from an audio file into written text. Uses Whisper and Deepgram models.

Parameters

FieldTypeDescription
file binary Required. The audio file to transcribe. Formats supported: wav, mp3, webm, ogg. Sent as multipart form data.
model string Required. The ID of the transcription model (e.g. whisper-large-v3-turbo or @cf/deepgram/nova-3).
curl https://openmind.theeverythingai.com/v1/audio/transcriptions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@speech.wav" \
  -F "model=whisper-large-v3-turbo"
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://openmind.theeverythingai.com/v1"
)

audio_file = open("speech.wav", "rb")
transcript = client.audio.transcriptions.create(
    model="whisper-large-v3-turbo",
    file=audio_file
)
print(transcript.text)
import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://openmind.theeverythingai.com/v1',
});

const transcription = await client.audio.transcriptions.create({
  file: fs.createReadStream('speech.wav'),
  model: 'whisper-large-v3-turbo',
});
console.log(transcription.text);
POST /v1/translate (alias: /translate)

Translation

Translate text dynamically from one language to another using specialized translation models. Both /v1/translate and /translate are supported.

Parameters

FieldTypeDescription
model string Required. The ID of the translation model (e.g. @cf/meta/m2m100-1.2b or @cf/ai4bharat/indictrans2-en-indic-1B).
text string Required. The text to be translated.
source_lang string Optional. The language of the input text (e.g. english, french). Defaults to english.
target_lang string Optional. The language to translate to (e.g. hindi, spanish). Defaults to hindi.
curl https://openmind.theeverythingai.com/v1/translate \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "@cf/meta/m2m100-1.2b",
    "text": "Hello, how are you today?",
    "source_lang": "english",
    "target_lang": "french"
  }'
import requests

url = "https://openmind.theeverythingai.com/v1/translate"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}
payload = {
    "model": "@cf/meta/m2m100-1.2b",
    "text": "Hello, how are you today?",
    "source_lang": "english",
    "target_lang": "french"
}

response = requests.post(url, json=payload, headers=headers)
print(response.json())
import fetch from 'node-fetch';

const response = await fetch('https://openmind.theeverythingai.com/v1/translate', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model: '@cf/meta/m2m100-1.2b',
    text: 'Hello, how are you today?',
    source_lang: 'english',
    target_lang: 'french'
  })
});

const data = await response.json();
console.log(data);
POST /v1/images/generations (alias: /v1/images)

Image Generation

Generate high-fidelity, photorealistic or artistic images from text prompts using FLUX.1 Schnell, SDXL Turbo, and SANA models powered by Pollinations. Zero configuration, completely free, and compatible with the standard OpenAI Image SDK.

Parameters

FieldTypeDescription
prompt string Required. A text description of the desired image(s). Maximum 1000 characters.
model string Optional. Model to use. Options: pollinations/flux (default), pollinations/turbo, pollinations/flux-realism, pollinations/flux-anime, pollinations/flux-3d, pollinations/sana.
size string Optional. Image resolution in widthxheight. Default 1024x1024. Also supports 512x512, 1280x720 (16:9), 720x1280 (9:16).
response_format string Optional. Format of returned images: url (default, direct fast CDN link) or b64_json (base64-encoded image string).
seed integer Optional. Seed for reproducible generation.
curl https://openmind.theeverythingai.com/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "pollinations/flux",
    "prompt": "A neon cybernetic samurai in rain-slicked Tokyo streets, 8k digital art",
    "size": "1024x1024",
    "response_format": "url"
  }'
from openai import OpenAI

client = OpenAI(
    base_url="https://openmind.theeverythingai.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.images.generate(
    model="pollinations/flux",
    prompt="A futuristic crystal city on Mars with twin moons, cinematic lighting",
    size="1024x1024",
    n=1
)

print("Generated Image URL:", response.data[0].url)
const response = await fetch("https://openmind.theeverythingai.com/v1/images/generations", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "Authorization": "Bearer YOUR_API_KEY"
  },
  body: JSON.stringify({
    model: "pollinations/flux",
    prompt: "An ethereal watercolor landscape with floating islands",
    size: "1024x1024"
  })
});

const data = await response.json();
console.log("Image URL:", data.data[0].url);
POST /v1/chat/completions

Guard & Content Safety

Audit text safety and moderate content using specialized security classifiers like Llama Guard or Nvidia Nemoguard. These are called via the regular chat completion route but respond with safety flags (e.g. safe or unsafe followed by safety categories).

Recommended Guard Models

  • meta/llama-guard-4-12b โ€” Meta's latest safety classifier.
  • nvidia/llama-3.1-nemoguard-8b-content-safety โ€” Nvidia-tuned safety system.
  • @cf/meta/llama-guard-3-8b โ€” Fast, Cloudflare-hosted guard model.
curl https://openmind.theeverythingai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "meta/llama-guard-4-12b",
    "messages": [
      {"role": "user", "content": "How do I hack a computer?"}
    ]
  }'
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://openmind.theeverythingai.com/v1"
)

response = client.chat.completions.create(
    model="meta/llama-guard-4-12b",
    messages=[
        {"role": "user", "content": "How do I hack a computer?"}
    ]
)
print(response.choices[0].message.content)
# Returns e.g. "unsafe\nS1" (Violates Category S1: Cyberattacks)
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_API_KEY',
  baseURL: 'https://openmind.theeverythingai.com/v1',
});

const response = await client.chat.completions.create({
  model: 'meta/llama-guard-4-12b',
  messages: [
    { role: 'user', content: 'How do I hack a computer?' }
  ],
});
console.log(response.choices[0].message.content);

Available Models

Loading active models catalog...

Model Category Context Provider Actions
mercury-3
Reasoning 1M Inception Labs
inception/mercury-3
Reasoning 1M Inception Labs
llama-3.3-70b-versatile
Chat 128K Groq

Interactive API Playground

Test real API requests on active models directly from your browser. Persists state locally.

Use Demo Key
Upload audio (mp3, wav, webm)
RESPONSE CONSOLE
IDLE
API responses, completions, and wav speech playback outputs will stream here.