API Documentation — ModelBridge

📖 API Documentation

ModelBridge provides a fully OpenAI-compatible API. If you've used the OpenAI SDK before, you already know how to use us — just change the base URL.

🚀 Quick Start

Get your first response in under 3 minutes. No credit card required.

1

Create a Free Account

Visit /auth/register.html and sign up with your email. No credit card needed.

2

Get Your API Key

After login, go to the Dashboard. Your API key (starts with mb-) is displayed at the top.

3

Make Your First Call

Copy the code below (Python / curl / Node.js) and replace YOUR_API_KEY with your actual key.

4

Check Usage

Visit the Dashboard anytime to view your token usage, remaining quota, and billing status.

First API Call — Copy & Paste

Python
curl
Node.js

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://aibridge-api.com/v1",
)

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
    

curl https://aibridge-api.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
    

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.API_KEY,
  baseURL: 'https://aibridge-api.com/v1',
});

const response = await client.chat.completions.create({
  model: 'deepseek-chat',
  messages: [{ role: 'user', content: 'Hello!' }],
});
console.log(response.choices[0].message.content);
    

✅ Expected response: a normal chat reply from DeepSeek V3. If you get 401, double-check your API key.

🔑 Authentication

How to Get Your API Key

Go to aibridge-api.com/auth/register.html and create a free account
After registration, you'll be redirected to the Dashboard
Your API key (format: mb-xxxxxxxxxxxxxxxxx) is displayed at the top of the dashboard
Click the copy button to copy it — you won't be able to see it again after leaving the page (but you can always generate a new one)

Using the API Key

All requests require an API key in the Authorization header:

Authorization: Bearer mb-xxxxxxxxxxxxx

⚠️ Security Note: Never expose your API key in client-side code (browsers). Always call the API from your backend server. If your key is compromised, generate a new one immediately from the Dashboard.

🌐 Base URL

https://aibridge-api.com/v1

🔗 Available Endpoints

Method	Endpoint	Description
GET	`/v1/models`	List available models
POST	`/v1/chat/completions`	Chat completion (supports streaming)
POST	`/v1/embeddings`	Generate text embeddings
GET	`/health`	Service health check

🤖 Supported Models

Model ID	Type	Context Window	Best For
`deepseek-chat`	Chat	64K	General purpose, coding (V3)
`deepseek-reasoner`	Reasoning	64K	Complex reasoning, math, logic
`deepseek-coder`	Coding	64K	Code generation & debugging (V2.5)
`deepseek-v4-pro`	Chat	128K	Flagship reasoning model, top-tier performance
`deepseek-v4-flash`	Chat	128K	Fast & lightweight, quick responses
`qwen-max`	Chat	32K	Multilingual tasks, long context
`qwen-plus`	Chat	131K	Cost-effective general usage
`qwen3-235b-a22b`	Chat	128K	Flagship Qwen3 model, best overall
`glm-4-plus`	Chat	128K	Advanced reasoning, complex tasks
`glm-4-air`	Chat	128K	Balanced performance & speed
`glm-4-flash`	Chat	128K	Fast & lightweight, cost-effective
`moonshot-v1-8k`	Chat	8K	Quick conversations
`moonshot-v1-32k`	Chat	32K	Medium context length
`moonshot-v1-128k`	Chat	128K	Long documents, deep analysis

💬 Chat Completions

The primary endpoint for generating text responses. Supports both regular and streaming (SSE) modes.

Request Parameters

Parameter	Type	Required	Description
model	string	Required	Model ID (e.g. `deepseek-chat`)
messages	array	Required	List of message objects. Each has `role` (`system`/`user`/`assistant`) and `content` (string)
temperature	float	Optional	Sampling temperature (0–2). Higher = more random. Default: 1.0
max_tokens	int	Optional	Max tokens to generate. Default varies by model
top_p	float	Optional	Nucleus sampling (0–1). Default: 1.0
stream	boolean	Optional	If `true`, returns SSE stream. Default: `false`
stop	string/array	Optional	Up to 4 sequences where the API will stop generating
presence_penalty	float	Optional	Penalty for repeated tokens (-2 to 2). Default: 0
frequency_penalty	float	Optional	Penalty based on token frequency (-2 to 2). Default: 0

Response Fields

Field	Type	Description
id	string	Unique completion ID (e.g. `chatcmpl-xxxxx`)
object	string	Always `"chat.completion"`
created	int	Unix timestamp of creation
model	string	Model ID used for the completion
choices	array	List of completion choices. Each has `index`, `message`, `finish_reason`
usage	object	Token usage: `prompt_tokens`, `completion_tokens`, `total_tokens`

Request Example

Python
curl

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user",   "content": "Explain quantum computing in 3 sentences."},
    ],
    temperature=0.7,
    max_tokens=500,
)
print(response.choices[0].message.content)
print("Tokens used:", response.usage.total_tokens)
    

curl https://aibridge-api.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "deepseek-chat",
    "messages": [
      {"role": "system", "content": "You are helpful."},
      {"role": "user", "content": "Hello!"}
    ],
    "temperature": 0.7,
    "max_tokens": 500
  }'
    

Streaming Response

Set stream: true to receive tokens as they are generated:

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Tell me a story."}],
    stream=True,
)

for chunk in response:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")
    

🔢 Embeddings

Generate vector embeddings from text. Compatible with OpenAI's /v1/embeddings endpoint.

Python
curl

response = client.embeddings.create(
    model="text-embedding-ada-002",
    input="The food was delicious",
)
print(response.data[0].embedding)  # List[float]
print(response.usage.prompt_tokens)
    

curl https://aibridge-api.com/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "text-embedding-ada-002",
    "input": "The food was delicious"
  }'
    

Note: Embeddings support varies by underlying model provider. Check the Supported Models table for availability.

⚡ Rate Limits

Plan	Requests / Min (RPM)	Tokens / Min (TPM)
Free	20	50,000
Pro	60	200,000
Custom	Custom	Custom

If you exceed rate limits, you'll receive a 429 error. Implement exponential backoff in production.

⚠️ Error Codes

Status Code	Error	Cause	Solution
401	Unauthorized	Invalid or missing API key	Check your Authorization header format: `Bearer mb-xxx`
404	Not Found	Invalid endpoint or model ID	Verify the URL and model name
429	Rate Limited	Too many requests	Reduce frequency; upgrade plan if needed
500	Internal Error	Upstream service issue	Retry after a few seconds
502	Bad Gateway	Upstream unavailable	Upstream AI provider is down
504	Gateway Timeout	Request took too long	Try with shorter prompts