ai.TokenHub

English

简体中文

English

简体中文

Appearance

FAQ ​

1. What is ai.TokenHub? ​

ai.TokenHub is an enterprise-level LLM Token platform that provides a unified API interface to access different AI models. You can use multiple models including GPT-4o, Claude, DeepSeek with a single API Key.

2. How to get started? ​

  1. Register and login to ai.TokenHub
  2. Create API Key in the console
  3. Use API Key to call endpoints

See Quick Start for details.

3. Which models are supported? ​

We support multiple mainstream AI models:

Chat Models:

  • OpenAI: GPT-4o, GPT-4o-mini
  • Anthropic: Claude-3-5-Sonnet, Claude-3-Opus
  • DeepSeek: DeepSeek-Chat

Embedding Models:

  • text-embedding-3-small
  • text-embedding-3-large

Rerank Models:

  • rerank-1, rerank-2

4. Is API compatible with OpenAI? ​

Yes, ai.TokenHub API is fully compatible with OpenAI API format. You can use OpenAI SDK directly, just modify base_url:

python
from openai import OpenAI

client = OpenAI(
    base_url="https://ai-tokenhub.com/v1",
    api_key="YOUR_TOKENHUB_KEY"
)

5. How does billing work? ​

  • Charged by token usage
  • Different models have different pricing
  • View real-time usage and costs in console
  • Supports recharge and quota management

6. Is there free quota? ​

New users receive a certain amount of free testing quota after registration for development and debugging.

7. How to handle rate limits? ​

When rate limit is exceeded, API returns 429 error. Recommendations:

  1. Implement retry mechanism, wait and retry
  2. Batch process requests to reduce request count
  3. Use streaming to reduce concurrency

See API Key Limits for details.

8. How to set fallback models? ​

Use fallback_models parameter:

json
{
  "model": "gpt-4o",
  "fallback_models": ["claude-3-5-sonnet", "deepseek-chat"]
}

System will automatically try fallback models when primary model is unavailable.

9. Is streaming supported? ​

Yes. Set stream: true to enable streaming. See Streaming for details.

10. How to get technical support? ​

  • Check Errors and Debugging documentation
  • Login to console to view usage logs
  • Contact customer service for help