Bonus: More April Fools pranks from Eiffel Tower Llama
Bonus: More April Fools pranks from Eiffel Tower Llama
You have landed upon a bonus post!
In bonus posts, I include extras as a thank-you to AI Weirdness supporters. It’s your financial support that helps me pay web hosting fees and other things I need to keep AI Weirdness running.
If you’re already an AI Weirdness supporter, sign in to see your bonus posts!
If you’re not an AI Weirdness supporter yet, you can subscribe and get access to this post, plus the entire archive of bonus posts. Also robot giraffes will sing your praises. If financial support isn’t in the cards right now, you can still become a free subscriber to get the regular AI Weirdness posts in your inbox.
This post is for subscribers only
Subscribe now
Already have an account? Sign in
Welcome back! You've successfully signed in.
Unable to sign you in. Please try again.
Success! Your account is fully activated, you now have access to all content.
Error! Stripe checkout failed.
Success! Your billing info is updated.
Error! Billing info update failed.
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
llama
TurboQuant seems to work very well on Gemma 4 — and separately, per-layer outlier-aware K quantization is beating current public fork results on Qwen PPL
I’ve been experimenting with TurboQuant KV cache quantization in llama.cpp (CPU + Metal) on Gemma 4 26B A4B-it Q4_K_M on an Apple M4 Pro 48GB, and the results look surprisingly strong. Gemma 4 findings On Gemma 4, QJL seems to work well, and FWHT as a structured rotation substitute also looks like a good fit for the large attention heads (dk=256/512). My benchmark results: tq3j/q4_0: 37/37 on quality tests, 8/8 on NIAH tq2j/q4_0: 36/37, with the only miss being an empty response +34% faster than q4_0/q4_0 at 131K context TurboQuant overtakes q4_0 from 4K context onward So on this setup, ~3.1 bits per K channel gets near-zero accuracy loss with a meaningful long-context speedup. What’s also interesting is that this looks better than the public Gemma 4 fork results I’ve seen so far. In the l
I built an AI fridge app that suggests Indian recipes before your food expires
The Problem I kept throwing away food because I forgot what was in my fridge. Sound familiar? What I Built FridgeSmart AI is a web app that: Tracks everything in your fridge and pantry Suggests Indian recipes based on what you already have Prioritizes ingredients that are about to expire Helps reduce food waste Tech Stack Frontend: React + Vite + TypeScript Backend: Node.js API Database: PostgreSQL (Neon) AI: Groq (Llama 3.3) Hosting: Render (free tier) Try It fridgesmart-ai-1.onrender.com Free to use — 3 recipe suggestions per day on the free plan. Would love your feedback!
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Models
How China is transforming Hong Kong into a strategic hub
Hong Kong’s first five-year plan is expected to guide the city’s future development. Never before has the city attempted a comprehensive plan in the style of mainland China, signalling a major shift in how it approaches long‑term growth. The real question is not why a laissez‑faire economy must adopt a new model but how this transformation will unfold. This exercise is unprecedented on multiple fronts. First, it departs from Hong Kong’s long-standing reliance on market forces and incremental...
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!