In this video, I discuss Meta’s recent release of the LLAMA 4 series of models and why it has generated significant controversy and backlash within the AI community. Despite promising big strides in AI with features like multimodal intelligence and colossal token contexts, the models face criticism for not meeting expectations. We compare the smaller LAMA 4 Scout, mid-sized LLAMA 4 Maverick, and the unreleased LLAMA 4 Behemoth against industry benchmarks. Moreover, reports suggest potential manipulations to inflate benchmark results, leading to severe credibility issues for Meta. I delve into community reactions, real-world test results, and industry opinion on the models’ actual performance. This episode uncovers why LLAMA 4’s launch has been deemed a tragic misstep for Meta.
▼ Link(s) From Today’s Video:
Official blog:
Huggingface Models:
Chubby’s Initial Excitement:
Pliny’s Llama 4 Jailbreak:
Jimmy Apples Vibe Check:
Flavio Coding Test:
EQ Bench:
Kalomaze’ Testing:
Meta Nerfed benchmarks Reddit post:
Misguided Attention Eval:
Bad Context Window testing:
Polyglot test:
Old “leak” from 2 months ago:
► MattVidPro Discord:
► Follow Me on Twitter:
► Buy me a Coffee!
————————————————-
▼ Extra Links of Interest:
General AI Playlist:
AI I use to edit videos:
Instagram: instagram.com/mattvidpro
Tiktok: tiktok.com/@mattvidpro
Gaming & Extras Channel:
Let’s work together!
– For brand & sponsorship inquiries:
– For all other business inquiries: [email protected]
Thanks for watching Matt Video Productions! I make all sorts of videos here on Youtube! Technology, Tutorials, and Reviews! Enjoy Your stay here, and subscribe!
All Suggestions, Thoughts And Comments Are Greatly Appreciated… Because I Actually Read Them.
00:00 Meta’s Weekend Blunder: Llama 4 Series Release
00:29 Diving into Llama 4 Models
03:31 Initial Community Reactions and Benchmarks
07:36 Vibe Checks and Community Tests
14:41 Allegations and Controversies
20:41 Concluding Thoughts on Llama 4
source
