Topic

Local AI

0 news articles · 3 tutorials

Tutorials

Gemma 3 4B vision running locally on a GTX 1660 for screenshot OCR, automation tutorial on AutoKaam

I Run Gemma 3 Vision On A 6GB GTX 1660 For Screenshot OCR: The Real VRAM And Latency Numbers

I host Gemma 3 4B vision on a single 6GB GTX 1660 for screenshot OCR and invoice extraction. Here are the install steps, the exact model tag, the VRAM it actually eats, and the cold versus warm latency I measured this week on my own desktop.

Jun 28, 2026·7 min read

Local speech-to-text stack on a 6 GB GTX 1660 with int8 Whisper, Silero VAD and pyannote diarization

⚡Automationadvanced

How I Run a Full Transcription Stack on a 6 GB GTX 1660

I transcribe my own audio locally on a 6 GB GTX 1660: int8 Whisper, Silero VAD, and speaker diarization, all under the VRAM ceiling. Here is the exact VRAM math, the int8 trade-off, and the capture pipeline that catches both my mic and the system output.

Jun 28, 2026·9 min read

⚡Automationadvanced

Running Gemma 4 Locally With Ollama, Setup Guide For Indian Devs

Local inference setup for Gemma 4 with Ollama on Mac, Windows, and Linux. Hardware requirements, performance benchmarks, and use cases where local beats cloud.

Apr 4, 2026·8 min read