Seapy retweeted
Apple just announced $22,299 for 512gb at 1.2 TB/s haa
An ~$18k RTX PRO 6000 + $1,000 used 3090 gives you 120GB at 1.8 TB/s and runs every model you'd actually use 2-3x faster.
The extra 392GB only lets you load models that decode at a dorky 3 tok/s.
You are actually paying thousands more to watch a progress bar.
vram > unimem
Seapy retweeted
There's a lot more action you can SEE from this stage than _____.
Bambu Lab H3C
Nov 3, 9 AM EST
Retractable rack. Full Visibility.
Sign up to get notified first: share.bambulab.com/4zzOIqB
Seapy retweeted
Save this post, the Bambu Lab Black Friday Playbook is here.
Beyond our best prices of the year and the 28 × H2C Combos giveaway on the official store, we're dropping a massive prize pool right here:
• 5 × Bambu Lab H3C (yes, our brand new H3C!)
• 20 × 4-Color Filament Bundles
• 50 × $50 Bambu Lab Store Coupons
• One Ultimate Grand Prize (revealed tomorrow)
You might want to clear some space in your workshop...
#BambuLab #BlackFriday #3DPrinting
Seapy retweeted
$TSLA
BREAKING: Turo and Tesla just announced the first verified Tesla FSD(Supervised) rental experience in the U.S., launching October 28.
• For guests: You can search for and book Teslas with FSD (Supervised) activated. FSD status is verified through a direct Tesla API integration.
• For hosts: Once a host links their Tesla account to Turo, FSD status is verified automatically and shown on eligible listings. Eligible U.S. hosts also get a free one-month FSD (Supervised) trial.
• Scale: Thousands of FSD-enabled Teslas are already on Turo. The Model 3 and Model Y are the 3rd and 4th most popular models on the platform this year.
• CEO quote: Turo CEO Andre Haddad says 55% of Americans, and 71% of Gen Z, are interested in self-driving tech for travel.
Seapy retweeted
[New Post] 3D 프린터로 만든 10인치 서버 랙
blog.outsider.ne.kr/1816
Seapy retweeted
Kimi K3.1 Pro 翻车了
MiniMax M3.1 Flash 翻车了
Muse wonder 翻车了
Gemini 4 Argon 也翻车了
只有 Deepseek V4.1 Flash 1017
大家边夸边用
Seapy retweeted
"The pace, and the expectations that came with it, felt almost as jarring [in China] compared with America as America's pace does compared with Europe. Everyone was just on it, all the time." world.hey.com/dhh/seven-year…
Seapy retweeted
Worst idea of the day goes to governmental official of relatively small country who suggested they would require AI companies to publicly disclose to their product safety commission within 24 hours of discovery all software vulnerabilities. 🤦🏻♂️🙅🏻♂️☠️
Seapy retweeted
🎉 MiniMax M3.1 Flash from @MiniMax_AI has Day-0 support in vLLM!
Built for multi-step automation, real-codebase engineering and visual reasoning—with native NVFP4 KV, NVFP4 MoE weights and DSpark speculative decoding in the serving stack.
What the model can do, and the vLLM engineering that makes it practical to serve. 👇
Model: huggingface.co/MiniMaxAI/Min…
1/5
MiniMax M3.1 Flash weights are now public. 428B parameters total, 23B activated per token.
Agentic performance: A generational leap in agentic tasks, coding, and cybersecurity — 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and 82.2 on CyberGym.
Efficient by design: MiniMax Sparse Attention and DSpark speculative decoding improve inference efficiency.
Hugging Face:
huggingface.co/MiniMaxAI/Min…
ModelScope:
modelscope.cn/models/MiniMax…
API: platform.minimax.io
M Plan: minimax.io/m-planMiniMax
MCode for desktop: agent.minimax.io/download
Seapy retweeted
As of today, during the Python Core Sprint, I have made 500 commits to CPython. It's been 910 days since my 400th commit. See you at 600 commits!
Thanks to all the core devs working with me. It's always such a pleasure to collaborate with you all!
Seapy retweeted
I remember when 200tps felt impossibly fast. This is going to be crazy.
DeepSeek V4.1 Flash 1017 runs at 1,521+ tok/s on Fireworks Ultrafast.
Fast enough to ship Tetris in 11 seconds, 7x faster than Claude Opus 5.5.
Join the Ultrafast waitlist: bit.ly/3UA0wKq
Seapy retweeted
Rejoice single DGX Spark owners!
You can now run Qwen3.8 Flash Next on TensorFold with the new Nvidia FP4 QAD quant 🔥
- 1.1M total FP8 KV cache
- 256k context, 5 concurrent users
- 46-55 tok/s on prose, single stream
- 132 tok/s on prose, 5 concurrent steams
- Prefill 2000-2300 tok/s
Qwen3.8 Flash Next is still the BEST model to run on one DGX Spark, bare none!
You can choose what to run: FP4 QAD for quality, or int4 for speed. Details on what is QAD below.
Get it here:
mia-ai.net/models/Qwen3.8-Fl…
Seapy retweeted
The 'old' vLLM recipe for DeepSeek v4.1 Flash *should* work with the new 1017 release. I haven't tried it though. Just try replacing the model.
That being said, I'm working on a TensorFold recipe.
vLLM recipe here:
github.com/MiaAI-Lab/DeepSee…
