viraldi.
Video details
How it works
Back to discovery
20:17
Technology & Computing / VIDEO

This Engine Makes 177B Qwen 3.8 Flash Extremely Fast with 12GB GPU | Strata

Codacus· 46.1K subscribers

28d ago · English audio

A voice agent placed a real grocery order, start to finish, in under two minutes. The model behind it was Qwen 3.8 Flash, 177B parameters counting its n-gram tables, running on one RTX 3060 with 12GB of VRAM and 64GB of system RAM. The engine that made that possible is Strata, an open-source (MIT) inference engine that is about two weeks old and runs exactly one model. llama.cpp supports around 150 architectures.…

Watch on YouTube
3.0×Score at discovery
95.4KViews
31.7KMedian · 34 videos

Outlier score

This score was measured on Tue, 03 Nov 2026 08:41:11 GMT. Current view counts may differ.

95.3K views ÷ a 31.7K baseline = 3.0×. Compared with 34 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: Tue, 03 Nov 2026 11:24:15 GMT. Counters may have changed.

Read the methodology

Generative AI & Agents

Channel-balanced percentile: 93

Comparison details

94 videos from 75 other channels, with each channel weighted equally. Same format and language, with 10,000–99,999 subscribers. Observed at 692 hours after publication, within ±48 hours.

This describes the collected sample, not all of YouTube. Updated Tue, 03 Nov 2026 12:00:41 GMT.

Niches & tags

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

Breakouts from other channels