TokenHub

What is TokenHub

TokenHub is an AI Model API Gateway and Marketplace that allows developers to compare major language, image, video, and speech models through a single OpenAI-compatible API. It provides a unified platform for accessing various AI models, enabling developers to route requests, compare provider pricing, and ship reliable AI products faster.

How to use TokenHub

1. Create a workspace: Sign up using your email address or Google account.
2. Fund shared usage: Add a balance to your workspace to cover model usage across supported providers.
3. Issue a gateway key: Generate an API key to integrate your application and route requests through TokenHub.

Features of TokenHub

• Unified API: Access a wide range of AI models (language, image, video, speech) through a single OpenAI-compatible API.
• Model Comparison: Compare pricing, capabilities, and availability of different AI models from various providers.
• Gateway-level Routing: Route traffic to the appropriate provider path while maintaining consistent request formats.
• Shared Usage Balance: Fund a single balance to cover usage across multiple providers and models.
• Multimodal Support: Route requests for text, vision, image, video, and speech models.
• Developer Controls: Standardize API keys, environments, billing, and vendor access for B2B integrations.

Use Cases of TokenHub

• AI Product Development: Quickly integrate various AI models into applications without managing multiple vendor-specific APIs.
• Model Evaluation: Compare different AI models for specific tasks (e.g., coding, research, image generation, speech synthesis) based on cost and performance.
• Cost Optimization: Route requests to the most cost-effective model for a given task.
• Streamlined Integrations: Simplify the process of integrating AI capabilities into B2B products.

Pricing

TokenHub offers a pay-as-you-go model with different pricing tiers based on usage. Specific pricing details are available for various models, including:

• GPT-6 Astra: Input: $10 / M tokens, Output: $50 / M tokens, Cache Read: $1 / M tokens
• Claude Fable 5.1: Input: $10 / M tokens, Output: $50 / M tokens, Cache Read: $0.25 / M tokens
• DeepSeek V4.1 Flash: Input: $0.2857 / M tokens, Output: $1.1429 / M tokens, Cache Read: $0.0057 / M tokens
• GLM-5.3: Input: $1.1429 / M tokens, Output: $4 / M tokens, Cache Read: $0.2857 / M tokens
• GLM-5.3-Flash: Input: $1.1429 / M tokens, Output: $4 / M tokens, Cache Read: $0.2857 / M tokens

Funding options include $20, $100, and $500.

FAQ

• What is the primary benefit of TokenHub? TokenHub allows developers to use one OpenAI-compatible API to access and compare a wide range of AI models, simplifying integration and management.
• What types of models are supported? TokenHub supports major language, image, video, and speech models from various providers.
• How is billing handled? A shared usage balance can be funded and used across different providers and models.