Flovo AI

Seamless Flow, Clear Outcomes

Standard View Circuit View

Local AI not prepared

click a branch to fold it · click a leaf to go deeper · drag to pan, scroll to zoom

Local AI models

Choose a local model. Qwen models are temporary by default and are not intentionally kept in Flovo's persistent model cache after the browser session. Edge/Max models are large installed models and stay until you remove them.

Selected modelNot downloaded Local storageModel-dependent InferenceOn-device WebGPU PromptsStay on-device

Qwen3 0.6B is the safest default. Qwen3 1.7B uses more memory. Gemma 4 Edge/Max use Google's LiteRT-LM WebGPU runtime and need much more disk and memory.