
One prepaid API for AI models

One prepaid API for AI models
Solo Router is an AI model router that provides a single API key and one prepaid balance for accessing multiple AI models from various providers. It supports models from DeepSeek, Z.ai (GLM), Moonshot (Kimi), Alibaba (Qwen), Meta, Google, MiniMax, NVIDIA, and others, including Llama and Gemma. Users pay per token at each model's listed price, using prepaid credit, with no subscription required. The service is designed to simplify the workflow of using multiple models by consolidating access and billing under one account. The platform offers per-token pricing, transparent cost reporting for each request, and the ability to set spending limits per API key. Prepaid credit can be topped up from $5, and new accounts receive free credit. Features include streaming via server-sent events, reasoning output, function calling, JSON mode, and support for images, video, voice, music, embeddings, and reranking. The service does not store prompts or responses; it logs only token counts and costs for billing purposes. Solo Router uses the standard chat-completions format, allowing existing code and tools such as LangChain, Cline, Continue, and most AI applications to work by changing the base URL to `https://solorouter.ai/api/v1` and using the provided key. Models can be switched by changing the model string, enabling workflows that route easier queries to lower-cost models and more complex questions to stronger models. It also supports building support bots, embedding documents for retrieval-augmented question answering, and generating images with models like Stable Diffusion, all using the same key and balance. The service also includes a browser-based chat interface for users who are not writing code, billed from the same prepaid balance. Account setup requires no forms or sales calls; users simply sign in, obtain an API key, and set the base URL. Typical use cases include using agent frameworks, building support bots, handling long-context sessions, and running production workloads without managing multiple provider accounts.