Sunny รีทวีตแล้ว
3つの異なる角度から撮影された動画では、一群の猫たちが古い井戸の周りを円を描くように一斉に動き、奇妙な儀式を行っている。
人間が撮影した映像、住宅の防犯カメラ、そしてクローズアップ映像――どの角度から見ても、これが合成やAIによるものではないことは明らかだ！
👇
日本語
Sunny
9.7K posts
Sunny
@sunnypause
I build the employees that never sleep, complain, or quit. AI agents for real businesses.
dark side of the moon เข้าร่วม Temmuz 2011
582 กำลังติดตาม297 ผู้ติดตาม
Sunny รีทวีตแล้ว
Qwen 3.5 27B is the release of 2026 so far
> Agentic model & great at tool calling
> Claude Sonnet 4.6 quality at home
> ~28GB in NVFP4
> Fits on a single RTX 5090
> with full context (256K)
Amazing model & performance
The prediction below will age like fine wine
Ahmad@TheAhmadOsman
Prediction We will have Claude Code + Opus 4.5 quality (not nerfed) models running locally at home on a single RTX PRO 6000 before the end of the year
English
Sunny รีทวีตแล้ว
btw this is the app that has generated $5000 in just 3 days, organically. it uses your Mac’s accelerometer sensor to moan when you slap it. we’re in the peak era of software.
anul agarwal@anulagarwal
You all are overthinking your app ideas. This app makes your Mac moan when you slap it -> it made $5,000 in 3 days. Just ship it.
English
Sunny รีทวีตแล้ว
Sunny รีทวีตแล้ว
🚨 THE DOMINOES ARE FALLING. IN DAYS, NOT WEEKS.
Here is the global oil crisis — country by country — as of today:
🇱🇰 Sri Lanka — RATIONING. 4-day work week. Schools closed.
🇵🇰 Pakistan — CRISIS. Overnight price surge. Long queues. 4-day work week.
🇮🇳 India — 9 DAYS of reserves left. Emergency suppliers being hunted.
🇰🇷 South Korea — 50 DAYS left. Clock ticking.
🇯🇵 Japan — LIED. Said 254 days. Actual usable reserves: 95.
🇬🇧 UK — Shell CEO warning: SHORTAGE STARTS IN APRIL.
🇩🇪 Germany — Gas up 30%. EU emergency plan launched.
🇫🇷 France — Paying 30% more at pumps than 8 weeks ago.
🇿🇦 South Africa — Government says "stable." Citizens photograph empty pumps.
🇹🇷 Turkey — Stocks crashed. Inflation exploding. Currency under pressure.
🇧🇷 Brazil — Watching nervously. Own oil helps but supply chains hurting.
🇦🇺 Australia — Tanker delays. Import dependent. Hoping IEA coordination holds.
🇺🇸 USA — Gas taxes suspended in states. SPR drawn down. Iran sanctions quietly paused.
🇨🇳 China — 1.4 BILLION barrels stockpiled. Banned exports. Still getting Iranian oil.
This is DAY 26 of the Hormuz blockade.
20% of global oil — GONE.
8 million barrels per day — GONE.
OPEC's response: +206,000 barrels. That is 2% of the hole.
Here is what nobody is telling you:
They're showing you "reserves are sufficient."
They're NOT showing you Japan overstated reserves by 3x, South Africa has empty pumps, and India has 9 days.
The real energy crisis hasn't even started yet.
RT so this doesn't get buried by the algorithm.
English
Sunny รีทวีตแล้ว
Sunny รีทวีตแล้ว
A young girl practicing martial arts while waiting for the bus. Sometimes, you need less than you think to stay active.
English
Sunny รีทวีตแล้ว
Sunny รีทวีตแล้ว
New on the Engineering Blog: How we designed Claude Code auto mode.
Many Claude Code users let Claude work without permission prompts. Auto mode is a safer middle ground: we built and tested classifiers that make approval decisions instead.
Read more: anthropic.com/engineering/cl…
English
Sunny รีทวีตแล้ว
I just hit my Claude Code 5 hour limit insanely fast.
What is happening?
English
Sunny รีทวีตแล้ว
Sunny รีทวีตแล้ว
Sunny รีทวีตแล้ว
Casual Bugatti Chiron sighting on Cambodian TikTok. Starting price around $3 million. Cambodia’s GDP per capita is about $2,800.
Via @monyy88887 on TikTok
English
Sunny รีทวีตแล้ว
Sunny รีทวีตแล้ว
first impressions of qwen 3.5 27B dense on a single RTX 3090.
35 tok/s. from 4K all the way to 300K+ context. no speed drop. hermes 4.3 started at 35 and degraded to 15 as context filled. qwen dense holds. MoE held 112 flat. 3x faster but only 3B of 35B active per token. architecture tradeoff.
Q4_K_M on 16.7GB. native context 262K. pushed past training limit to 376K before VRAM ceiling on 24GB. tried q8 KV cache at 262K, speed collapsed to 11 tok/s. q4_0 KV is the sweet spot. flash attention mandatory.
built in reasoning mode. the model thinks step by step before it answers. full chain of thought surviving Q4 quant. 1,799+ token thinking chains with self correction loops. on a single consumer GPU.
gave it one prompt: "build a realtime particle galaxy simulation in one HTML file." 3,340 tokens. 95 seconds. one shot. ran on first load. full reasoning and coding in the video below.
optimal config if you want to skip the hours of testing:
llama-server -ngl 99 -c 262144 -fa on --cache-type-k q4_0 --cache-type-v q4_0
this is just the warmup. octopus invaders is next: 10 files, 3,400+ lines, zero steering. the prompt hermes quit at 22%.
already more impressed than expected. full results coming soon.
Sudo su@sudoingX
last time this qwen 3.5 MoE one shotted a full space shooter game. 3,483 lines across 10 files. ran on first load. zero steering. 112 tok/s on a single 3090. then i ran the same prompt on hermes 4.3 36B dense. similar size model, completely different architecture. it wrote 1,249 lines, declared done with empty files, needed three steering interventions, and the game didn't work. used 22% of available context and quit. nine posts and two GPU configs later the conclusion was clear: the bottleneck wasn't hardware. but that leaves a question. was that a dense architecture problem or a hermes 4.3 problem? qwen is the only family that ships both. 35B MoE with 3B active per token. and a 27B dense with all 27B active per token. same team. same training pipeline. different architecture. downloading qwen 3.5 27B dense now. Q4_K_M. same quant. same single RTX 3090. same octopus invaders prompt. if it finishes the game clean, hermes was the problem. if it fails the same way, dense architecture doesn't have the endurance for autonomous coding on consumer hardware regardless of who builds it. the tiebreaker.
English