loaf
11.3K posts

loaf
@lordOfAFew
dreaming up the future: @daydreamsagents terraforming onchain worlds: @cartridge_gg @LootRealms @ohayo_dojo @realms_gg

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI

🔮 SpaceX IPO April 20, 2026.


I got a 1T (trillion) parameter model running on my MacBook Pro. Kimi-K2. 1.029T params. ~1 TB raw weights. 524 GB converted. ~1.7 tok/s. Yesterday it was 671B. Today it's 1T. Same laptop. Same M4 Max. No cloud. When I say we: I mean Claude and me.


We've just introduced laws to crack down on petrol price gouging with huge fines.




API Server with Responses API Hermes can now act as an OpenAI-compatible backend — any frontend (Open WebUI, LobeChat, LibreChat, ChatBox, etc.) can connect to it. Exposes both /v1/chat/completions and /v1/responses (stateful, with previous_response_id chaining). Full agent stack behind the API: tools, skills, memory, cron.




Love Island is the U.S.’s #1 reality show - $30 million budget, ~10 million viewers / episode In less than a week, an AI version of Love Island starring fruit has racked up 15 million viewers / episode 🤯 And, it surpassed the real Love Island in followers today (3.3 million)


testing cascade 2 on a single 3090 right now. same card i tested qwen 3.5 35B-A3B on at 112 tok/s. same active params, same VRAM tier, different hybrid architectures. mamba vs deltanet head to head. numbers coming tonight. if a spark lands on my desk next you'll get those numbers too.





