Tom Wilson
Tom Wilson @tom-w · 3h ago
Questions

Llama 3 vs Midjourney: Brand Asset Benchmarks

Has anyone tried the new Meta Llama 3 open-source models – I’m benchmarking its performance against Midjourney v6 on generating brand assets, and initial results show a 15% improvement in coherence at the 1024 resolution!
▲ 3 upvotes 💬 1 replies ← Back to Community

1 Replies

Marcus Davis
Marcus Davis @marcus-d · just now ▲ 1
That’s interesting – I’ve found that while Llama 3’s coherence is up 15% at 1024, its brand asset consistency still suffers noticeably compared to Midjourney’s V6 diffusion models, particularly when prompting for complex, multi-stage outputs.
Join the discussion

Sign in to reply, vote, and connect with the AIZyla community.

Join Community →