xiangsx/gpt4free-ts

Features

• OpenAI-Compatible APIs - Exposes a standardized API that mimics the OpenAI chat completions JSON schema for broad tool compatibility.
• Free AI API Endpoints - Provides a unified interface for accessing large language models via free API endpoints.
• AI Model APIs - Provides a unified interface to interact with diverse AI models without requiring paid subscriptions.
• Model Request Routing - Routes incoming prompts to various third-party AI providers based on availability and priority maps.
• AI Provider Integrations - Integrates multiple free AI model providers into a single programmatic TypeScript interface.
• Chat Completion Services - Provides a standardized interface for generating conversational chat completions using structured message sequences.
• Conversation State Management - Maintains conversational context by injecting previous message history into prompts for stateless provider endpoints.
• LLM Completion Requests - Enables requests for LLM text completions via a free interface to support agentic workflows.
• Generative Text Inference - Enables the retrieval of intelligent text responses from large language models via a unified interface.
• Language Model Response Generators - Wraps multiple free provider endpoints to generate and retrieve responses from various large language models.
• LLM Gateways - Functions as a TypeScript-based centralized gateway for aggregating multiple free LLM providers.
• Model Provider Management - Orchestrates and manages the target third-party websites used to host accessible language models.
• Traffic Load Balancers - Distributes incoming requests across multiple AI provider sites using a priority map for high availability.
• AI Model Load Balancers - Balances request loads across various third-party AI providers to prevent service interruptions.
• AI Provider Proxies - Proxies requests to free third-party AI providers to allow model access without paid subscriptions.
• AI Prompt Sending APIs - Provides a programmatic API to transmit prompts and receive completions from LLMs without direct provider payment.
• Multi-Provider Model Aggregation - Allows sending prompts to various third-party AI websites through a single unified interface.
• Server-Sent Events - Implements server-sent events to deliver AI-generated text chunks to the client in real time.
• OpenAI-Compatible API Servers - Mimics the OpenAI chat completions format to integrate free AI models into existing tools.
• Streaming Chat Responses - Delivers AI-generated chat responses incrementally via event streams for real-time user interaction.
• Chat History APIs - Manages the injection of conversational history into prompts to provide context for new responses.
• Interchangeable Model Support - Provides a mechanism to list supported models and the specific hosting sites that offer them.
• LLM Response Streaming - Implements token-by-token incremental delivery of large language model outputs to the user.
• Model Availability Discovery - Allows users to list available provider sites and supported models to determine current compatibility.
• Model Response Streaming - Provides real-time incremental delivery of language model text responses via event streams.
• Sensitive Word Filters - Scans and modifies user prompts to remove prohibited or sensitive words before processing.
• Real-Time Text Streaming - Implements low-latency event streams for the real-time delivery of LLM token streams.
• AI Server Containerization - Packages the AI gateway and its API interfaces into a Docker container for portable hosting.
• Request Token Validators - Validates secret tokens in request headers to restrict access to the API gateway.
• Request Interception Middleware - Utilizes request interception middleware to verify secret tokens and filter prohibited content.

Star history