viraldi.
Video details
How it works
Back to discovery
7:14
Technology & Computing / VIDEO

A 125B Local Model on a 16GB GPU: Fast, But Is It Smart?

Zero to MVP· 33.3K subscribers

19d ago · en-US audio

Can a 125-billion-parameter model run on a regular 16GB GPU and still do useful work? I ran Qwen3.8-Flash-Next with Strata on an RTX 4060 Ti, measured generation speed, and gave it three coding tasks of increasing difficulty using Pi as the agent. 🖥️ Setup CPU: AMD Ryzen 7 (https://link.amazon/B07UXubRX) GPU: RTX 4060 Ti 16GB (https://link.amazon/B0c6ku9nm)…

Watch on YouTube
3.0×Score at discovery
69.2KViews
22.9KMedian · 39 videos

Outlier score

This score was measured on Sun, 25 Oct 2026 08:06:21 GMT. Current view counts may differ.

69.2K views ÷ a 22.9K baseline = 3.0×. Compared with 39 earlier videos from the same channel, matching the format recorded at scoring time. Older-upload baseline: reference videos had at least as long to gain views. This comparison is not age-matched.

YouTube snapshot: Sun, 25 Oct 2026 10:22:48 GMT. Counters may have changed.

Read the methodology

Artificial Intelligence & Data

Channel-balanced percentile: 100

Comparison details

21 videos from 17 other channels, with each channel weighted equally. Same format and language, with 10,000–99,999 subscribers. Observed at 471 hours after publication, within ±48 hours.

The comparison uses this broader topic because the specific niche has too few examples. This describes the collected sample, not all of YouTube. Updated Sun, 25 Oct 2026 10:36:23 GMT.

Niches & tags

Transcripts and detailed analytics are available for recent outlier Shorts up to three minutes. Research tools

Breakouts from other channels