Generative models, prompting, agents and AI tool capabilities.
LFM3 on Ollama: 20 INSANE Real Tests (1 GB Model, 141 tok/s, 1 Epic FAIL)
3d ago ยท English audio
I ran 20 real tests on LFM3 (1.7B) through Ollama on an RTX 4060 laptop: easy to hard, real outputs, executed code, real SQL, validated JSON. Result: 16 PASS, 3 PARTIAL, 1 FAIL. About 141 tokens/sec in roughly 1 GB, but it also hallucinated a paper that does not exist. Install: ollama.com/download, needs Ollama 0.40.13 or newer. Then: ollama pull lfm3 0:00 Introโฆ
Prompt Engineer 48
๐ง The AI revolution isn't waiting. Neither should you. Welcome to PromptEngineer48 โ where we master the language of AI so you don't have to figure it out alone. This channel is your go-to hub for staying ahead in the world of Artificial Intelligence. From crafting the perfect prompt to deploying full AI-powered workflows, we cover it all.โฆ
Outlier score
This score was measured on Wed, 04 Nov 2026 17:29:50 GMT. Current view counts may differ.
3K views รท a 704 baseline = 4.3ร. Compared with 32 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: Wed, 04 Nov 2026 20:08:07 GMT. Counters may have changed.
Read the methodologyGenerative AI & Agents
Channel-balanced percentile: 48
Comparison details
50 videos from 41 other channels, with each channel weighted equally. Same format and language, with 10,000โ99,999 subscribers. Observed at 94 hours after publication, within ยฑ19 hours.
This describes the collected sample, not all of YouTube. Updated Wed, 04 Nov 2026 21:09:42 GMT.
Niches & tags
Transcripts and detailed analytics are available for recent outlier Shorts up to three minutes. Research tools