cheahjs/free-llm-api-resources

Free LLM API resources is a curated, regularly updated list of services that let you access large language models through an API for free, or with free trial credits. It is aimed at people building AI-powered apps or experimenting with models like Llama, Gemini, and other open or proprietary systems, who want to avoid paying for API access while testing or building something small. The README explicitly asks people not to abuse these free tiers, since that risks the community losing access to them, and it excludes any service that works by reverse engineering an existing chatbot rather than offering a legitimate API. The list is split into two main sections: providers offering ongoing free access, and providers offering trial credits that eventually run out. The free-access section covers services such as OpenRouter, Google AI Studio, NVIDIA NIM, Mistral, HuggingFace Inference Providers, Vercel AI Gateway, Cerebras, Groq, Cohere, GitHub Models, and Cloudflare Workers AI. For each provider, the README lists the specific models available and the rate limits attached to them, such as requests per minute, requests per day, or tokens per minute, since these limits vary a lot between providers and change over time. The trial-credit section lists services like Fireworks, Baseten, Nebius, Novita, AI21, Upstage, NLP Cloud, Alibaba Cloud's Model Studio, Modal, Inference.net, Hyperbolic, SambaNova Cloud, and Scaleway, which typically give a fixed amount of free credit or trial period rather than an indefinitely free tier. Some providers require phone number verification or opting into having your data used for model training as a condition of the free tier, and the README notes this where it applies. The README states it is generated automatically by a script in the repository, so the list of models and limits stays current with what each provider is offering at any given time rather than being manually maintained by hand.