Frequently Asked Questions About AI Model APIs

An AI model API is a programming interface that lets you send text (or images, audio, video) to an AI model and get a response back. It's the easiest way to add AI capabilities to your application — you send a request, the AI processes it, and returns an answer. No need to run your own GPUs or manage complex infrastructure.

ModelHub provides a single API that gives you access to 45+ AI models from providers like DeepSeek, OpenAI, Anthropic, Google, and more. Learn more in our docs.

Getting started takes less than 30 seconds:

1. Go to your dashboard
2. Enter your email address (no credit card needed)
3. Copy your API key — you get $5 free credit instantly
4. Start making API calls using the OpenAI-compatible endpoint at https://modelhub-api.com/v1

See our beginner's guide for a step-by-step walkthrough.

No. ModelHub gives you $5 in free credit just for signing up with your email. No credit card required. You only need to add a payment method when you're ready to purchase additional credits or a subscription plan.

No. ModelHub is built for international developers. You sign up with your email, and you can pay with international credit cards or Paddle (which supports PayPal, Visa, Mastercard, and more). No Chinese phone number needed.

This is one of the key advantages of using ModelHub over DeepSeek's direct API. See our guide on getting a DeepSeek API key without a Chinese number for more details.

ModelHub uses a pay-per-token pricing model. You're charged based on the number of tokens (roughly 0.75 words per token) processed by each model. Different models have different per-token prices.

For example, DeepSeek V4 Flash costs $0.14 per million input tokens and $0.28 per million output tokens. You can also choose subscription plans for predictable monthly billing.

See our pricing page and pricing guide for full details.

Yes, significantly. ModelHub's DeepSeek V4 Flash costs $0.14/M input tokens compared to OpenAI's GPT-4o at $2.50/M. For output, DeepSeek V4 Flash is $0.28/M vs GPT-4o's $10.00/M. That's up to 36x cheaper.

Even compared to OpenAI's budget option (GPT-4o Mini at $0.15/M input), ModelHub is more affordable and typically delivers better quality. See our AI API cost comparison for detailed breakdowns.

ModelHub uses Paddle for payment processing, which accepts:

• Visa, Mastercard, American Express
• PayPal
• Apple Pay
• Google Pay
• And other regional payment methods

Payments are processed in USD. All major international cards are supported.

Yes. Your dashboard shows real-time token usage, costs, and remaining balance. You can track usage across all 45+ models, set budget alerts, and view historical usage patterns.

Yes. ModelHub is fully compatible with the OpenAI Python, Node.js, Go, and Java SDKs. To switch from OpenAI to ModelHub, you only need to change:

1. The base URL from https://api.openai.com/v1 to https://modelhub-api.com/v1
2. Your API key to your ModelHub key
3. The model name (e.g., from gpt-4 to deepseek-v4-flash)

Everything else — message format, streaming, function calling, embeddings — works identically. See our migration guide for step-by-step instructions.

ModelHub provides access to 45+ AI models including:

• DeepSeek: V4 Flash, V4 Pro, Reasoner, Coder, Embeddings
• Anthropic: Claude Sonnet 4, Claude Haiku 3.5
• OpenAI: GPT-4o, GPT-4o Mini, GPT-5.5, o3 series
• Google: Gemini 2.5 Flash, Gemini 2.5 Pro
• Mistral: Mistral Large 3, Mistral Small
• Meta: Llama 4 series
• And more: Qwen, Yi, Cohere, and community models

See the full list on our pricing page.

Yes. ModelHub supports Server-Sent Events (SSE) streaming for all chat completion models. Use stream: true in your request, and you'll receive tokens in real-time using the exact same format as OpenAI's streaming API.

Yes. Function calling (tool use) is fully supported across all major models. The tools and tool_choice parameters work as documented in the OpenAI API reference. DeepSeek V4 Flash, Claude Sonnet 4, and GPT-4o all handle function calls effectively.

Context window sizes vary by model:

• DeepSeek V4 Flash: 128K tokens
• Claude Sonnet 4: 200K tokens
• Gemini 2.5 Pro: 1M-2M tokens
• GPT-4o: 128K tokens

Choose the model with the context window that fits your use case. Available all through one API key on ModelHub.

ModelHub is a unified AI API platform that gives you access to 45+ AI models from multiple providers through a single API key and compatible OpenAI-format endpoint. Think of it as a universal adapter for AI models.

Instead of managing separate accounts with DeepSeek, OpenAI, Anthropic, Google, Mistral, and others — you get one account, one API key, one bill, and one consistent API format. Switch between models by changing a single parameter.

There are several key differences:

• No Chinese phone number needed — sign up with just your email
• International payment methods — credit cards, PayPal via Paddle
• 45+ models — not just DeepSeek, but also Claude, GPT, Gemini, and more
• OpenAI-compatible API — your existing code works with a base URL change
• Dashboard — track usage, monitor costs, manage API keys
• Free $5 credit — no credit card required to start

Read our DeepSeek vs OpenAI pricing guide for a deeper comparison.

The fastest way to get help is to use the AI Assistant (the robot icon at the bottom-right of every page). It can answer most questions about pricing, setup, and technical issues instantly.

For specific account or billing issues, reach out via the contact form on the dashboard.