Cliyer AI

Cliyer AI is an AI model runner that lets developers chat with any open-source Ollama model through a managed cloud interface, without setting up local infrastructure. It targets machine learning engineers, full-stack developers, and AI practitioners who want on-demand access to open-source models without managing their own hardware.

The platform connects to the Ollama model library, giving users access to hundreds of specialized open-source models including Llama, Qwen, Mistral, and DeepSeek. Users can pull any model by entering its tag directly through the Model Manager, and switching between models mid-conversation is supported without restarting a session. Cliyer AI runs inference on enterprise-grade engine clusters, so there is no local GPU required and no configuration overhead.

Billing is based on actual inference time in seconds rather than a flat monthly subscription. Three credit tiers are available: a free tier with 100 credits supporting models under 10GB on the Starter Engine, a $5 tier with 1,000 credits supporting models under 40GB with priority infrastructure, and a $25 tier with 5,000 credits unlocking all engine tiers and models up to 60GB. The platform currently supports models up to 70GB in size. For teams evaluating multiple open-source AI models, this AI model runner removes the barrier of infrastructure setup while keeping costs tied directly to usage.

Cliyer AI is primarily used for chatting with open-source AI models from the Ollama library through a cloud-hosted interface, removing the need for local GPU setup. Developers and ML teams also use it to compare outputs across multiple models such as Llama, Qwen, and Mistral within a single session.