Text-Generator is a self-hostable AI text generation API platform designed to let developers access leading AI models through a familiar API, covering capabilities such as chat, code, reasoning, and vision.
Main capabilities:
- Unified API: Provides an OpenAI-compatible interface, so developers can point the OpenAI client to Text Generator without rewriting their application, thereby accessing a network of multiple model providers.
- Model selection: Supports manually selecting a specific model, or using Auto mode to automatically match a suitable model for each request, balancing speed, quality, and cost.
- Built-in fallback: When a provider is rate-limited or unavailable, requests can still continue to be processed.
- Unified balance: A single top-up can be used for all routed models, with usage managed centrally in one account.
- Key management: You can use the hosted gateway, or keep the option of using your own provider keys.
- Privacy by design: The model selection process does not send prompts to another AI call.
- Multi-surface use: Provides UI chat, playground, and built-in tools, as well as API calling methods.
- Open source: The complete platform code from the API to the router is available on GitHub, supporting running it yourself and participating in improvements.
How to use:
- After subscribing, you can use chat, UI tools, and provider APIs without limits; you can also start for free and add API credits as needed.
- Call through the OpenAI-compatible API, supporting chat completions and streaming responses, aliases, Auto configuration and exact model IDs, as well as routing metadata and provider response headers.
- Can be self-hosted, inspect the code, and run it.
Use cases:
- Developers who need to build AI features with controllable cost and a privacy-first approach.
- Applications that want unified access to multiple model providers and reduced integration effort.
- Projects that need multimodal capabilities such as chat, code, reasoning, and vision.
Limitations:
- The official text does not provide clear explanations regarding specific implementation details, pricing, or platform support for crawling, image analysis, and speech-to-text.
