A browser-based demo of the LFM2.5-VL-1.6B vision-language model using WebGPU acceleration. Process video streams and generate captions entirely on your device with no server communication. Supports adjustable input resolutions and quantization options.

Vision language modelWebcam processingWebgpuReal time captions

Checked by Vibed

Secure connection (HTTPS)
Runs right here in Vibed
Works on phones
Gets no camera or microphone access in Vibed
Private: Vibed doesn’t see what you type in it

liquidai-lfm2-5-vl-1-6b-webgpu.hf.space ↗Found on Hugging Face ↗

More to explore

AZN8A SaaS and startup directory built for AI agents and makers.AI ToolsOpen

The Novel RoomA private literary studio for discovering, writing, editing, and publishing books.AI ToolsOpen

Interview VIPThe #1, best, and fastest desktop AI interview assistant for live guidance.AI ToolsOpen

Anomaly AIAI data analyst for large datasets and spreadsheetsAI ToolsOpen

Sora2 HubAI image and video generation platform. Create videos and images from text or images with multiple leading models, in HDAI ToolsOpen

Interview ManThe #1, best, and fastest AI interview assistant for real-time guidance.AI ToolsOpen

Made a web app?List it on Vibed for free. It goes live right away.

Submit your app

Free

Runs right here. No download.

Save

Checked by Vibed

Secure connection (HTTPS)
Runs right here in Vibed
Works on phones
Gets no camera or microphone access in Vibed
Private: Vibed doesn’t see what you type in it

liquidai-lfm2-5-vl-1-6b-webgpu.hf.space Found on Hugging Face