Welcome to the Token1.US FAQ. Here you'll find answers to the most common questions about our unified AI API gateway, pricing, models, security, and technical integration. Can't find what you're looking for? Contact us.
Token1.US is a unified AI API gateway that provides a single OpenAI-compatible endpoint for accessing multiple AI models including GPT-4o, Claude 3.5 Sonnet, DeepSeek-V3, and GLM-4-Plus. Developers use one API key and one SDK to call any model, with transparent per-token billing. Instead of managing multiple API accounts, keys, and SDKs across different providers, you get everything through one unified interface.
Sign up at discount-token.com, navigate to API Keys in your dashboard, and generate a new key. Your key works immediately across all supported models. New accounts may receive free trial credits to test the platform.
Yes, Token1.US is fully OpenAI-compatible. Simply change the base_url to https://discount-token.com/v1 and use your Token1.US API key. All existing OpenAI SDK code — whether Python, JavaScript, Go, or any other language — works without modification. See our migration guide for details.
from openai import OpenAI
# Before (OpenAI direct)
# client = OpenAI(api_key="sk-...", base_url="https://api.openai.com/v1")
# After (Token1.US) - just change base_url and api_key
client = OpenAI(api_key="your-token1-key", base_url="https://discount-token.com/v1")
response = client.chat.completions.create(
model="gpt-4o", # or "deepseek-v3", "claude-3-5-sonnet", "glm-4-plus"
messages=[{"role": "user", "content": "Hello!"}]
)
Token1.US charges transparent per-token pricing with zero markup. You pay the exact provider rate. Costs depend entirely on which model you choose:
Use our API cost calculator to estimate your monthly spending, or read our detailed token pricing guide.
Billing is calculated per request based on actual token consumption. Input and output tokens are billed separately at the provider's rate. There are no monthly fees, no markup, and no hidden charges. You see real-time usage and costs in your dashboard. For a deeper understanding of how tokens work, see our token pricing explained guide.
Yes. You can set daily, weekly, or monthly spending limits in your dashboard. When a limit is reached, API requests are paused and you receive an alert. This prevents unexpected charges from bugs, traffic spikes, or infinite loops in your code.
Input tokens are the text you send to the API (your prompt, system message, conversation history). Output tokens are the text the model generates in response. Output tokens typically cost 2-5x more than input tokens because text generation requires more computational resources than text processing.
Yes. One Token1.US API key gives you access to GPT-4o, Claude 3.5 Sonnet, DeepSeek-V3, GLM-4-Plus, and more. Switch models by changing the model parameter in your request. No separate accounts or keys needed. Compare all models in our model comparison or read our GPT-4o vs Claude deep-dive.
It depends on your use case:
| Use Case | Recommended Model | Cost |
|---|---|---|
| Cost-sensitive bulk processing | DeepSeek-V3 | $0.14/$0.28 |
| General-purpose chat/Q&A | GLM-4-Plus | $0.70/$0.70 |
| Complex reasoning, multimodal | GPT-4o | $2.50/$10.00 |
| Coding, nuanced writing | Claude 3.5 Sonnet | $3.00/$15.00 |
Yes. Function calling (tool use) is supported across GPT-4o, Claude 3.5 Sonnet, and other models that support it. The API format is identical to OpenAI's function calling specification. Learn more in our developer guide.
Yes. All models support streaming (Server-Sent Events). Set stream=true in your request to receive tokens incrementally. This dramatically reduces perceived latency for chatbots and interactive applications. For a comparison of streaming vs batch modes, see our streaming vs batch guide.
Context windows vary by model: Claude 3.5 Sonnet supports 200K tokens, GPT-4o supports 128K tokens, GLM-4-Plus supports 128K tokens, and DeepSeek-V3 supports 64K tokens. Choose the model with sufficient context for your use case.
Token1.US aggregates rate limits across all models. When one model hits its limit, requests automatically failover to alternative models. This gives you effectively unlimited throughput as long as at least one model is available. Rate limit headers are included in every response for monitoring. See our rate limiting guide for details.
Token1.US automatically fails over to alternative models when one provider is unavailable. Your requests continue working without interruption. The system includes circuit breaker protection, automatic retries with exponential backoff, and transparent error reporting. Read our error handling guide for implementation details.
Always implement retry logic for transient errors (429, 500, 502, 503, 504) using exponential backoff with jitter. Don't retry client errors (400, 401, 403). Token1.US handles many error scenarios automatically, but your code should still have fallback logic. See our comprehensive error handling guide with code examples.
Yes. API keys are hashed with SHA-256 and never logged in plaintext. All communication uses HTTPS/TLS encryption. Token1.US does not store your request content or use it for model training. Data is processed in-memory and discarded after response delivery. We add security headers including HSTS, X-Frame-Options, and Content-Security-Policy.
No. Token1.US does not use your request data, prompts, or generated content for training any models. Your data is processed to fulfill your API request and then discarded. We do not retain conversation logs or prompt history beyond what's needed for billing reconciliation.
Yes, migration takes one line of code. Change your base_url from api.openai.com/v1 to discount-token.com/v1 and replace your API key. All your existing code, SDK calls, function definitions, and prompt templates work unchanged. See our step-by-step migration guide or explore OpenAI API alternatives.
Yes. Token1.US is designed for commercial use with no restrictions. The content you generate belongs to you. Review our terms of service for full details. Explore AI API use cases for inspiration.
Support is available through the dashboard, email, and community channels. Technical questions are typically answered within 24 hours. API status and incident reports are available in real-time on the about page.
Still have questions? Contact us or get started with free trial credits.