Where art meets intelligence. Building human-amplified digital ecosystems. Founded by @malekoo.
Framingham, MA United States
Joined December 2021
- Tweets106
- Following37
- Followers89
- Likes201
AIC retweeted
Thanks to @bot and the vision of @elonmusk, a big shift in ecommerce is coming.
Tonight Grok Bot pulled up beautiful shopping cards while I was shopping for olive oil, then connected to my Shopify store, rewrote my first listing, audited the whole site and built a full redesign on a draft theme, all from a chat.
When one door closes, many more open!
AIC retweeted
Apple just put touch on the MacBook Pro, then had to redesign the hinge so the screen wouldn't wobble when you tap it. That tells you everything.
Touch is powerful. Put it on a machine this capable and you get something incredible: the intimacy of an iPad with the muscle of a Pro. But a screen bolted to a keyboard is a screen you have to reach for. Your arm floats. Your wrist tires. The display can only sit where the hinge lets it. That's not freedom. That's a compromise.
What I want is a display that lets go. One you hold in your hands like an iPad, lay flat to sketch, or tilt on your desk to work. Maybe the breakthrough isn't the laptop at all. Maybe it's the display next to it.
Free the screen. Then touch makes sense.
AIC retweeted
Just ordered: Mac Studio with M5 Ultra and 512 GB of unified memory.
36-core CPU, 80-core GPU with Neural Accelerators in every core, 1.2 TB/s memory bandwidth. Apple rates it at up to 4.3x the peak AI compute of M3 Ultra.
A little pricey, but a great investment in our lab to keep pushing local AI research and development.
It joins our two M3 Ultras (512 GB each, 819 GB/s) over Thunderbolt 5. That's 1.5 TB of unified memory in the room for open frontier models. Fully local. Tokens never leave.
Arriving in 3 to 4 weeks. Head-to-head benchmarks against the M3 Ultras as soon as it's racked.
AIC retweeted
Great to see the collab between Meta and Apple pushing MLX and local AI forward. The underrated part: MLX loads Meta's official NVFP4 checkpoint straight from Hugging Face, no conversion needed, with RDMA over Thunderbolt 5, efficient MoE support and the M5 GPU Neural Accelerators. Testing next on our 2x M3 Ultra Mac Studios (80-core GPU, 512 GB each).
Muse Spark Open -- weights releasing today -- runs great on new Mac Studios thanks to work with @johnternus and Apple.
AIC retweeted
First @typesafeai use case, live in our Mac app: setup and troubleshooting help when no model is loaded.
Model downloading, load failed, API returning 503, phone won't pair: the user asks, Jev reads the question with the whole built-in manual as state and decides, with probabilities, what it is and which article answers it, or that nothing does. The app then shows the real documentation and live status. Jev decides, the app answers from its own docs. No model loaded, nothing invented.
42/42 on a held-out set: paraphrases, typos, French, German, Spanish, features that don't exist, follow-ups. Median 0.93 s.
Great breakthrough by the TypeSafe team. Thank you.
AIC retweeted
Prompt cache on our headless dual Mac Studio cluster was stuck at 0%. Silent anomaly.
Agent turns re-prefilled 12k–29k tokens every step (56% of a 10 min session).
Two server fixes → 99/100/100% hits, 18.5 s → 12.0 s per turn.
Fix #1: make the cache key stable across tool loops. Fix #2: a one-token warm so the next turn actually hits.
If your local agent feels mysteriously slow, check cache hits before you blame the hardware.
AIC retweeted
There's growing talk, in more than one capital, of limiting who can download open source model weights. It's like telling early developers that programming languages were too dangerous to share. We shared them anyway, and it built the modern world. Labs made the breakthroughs. Home desks will carry them further. History repeats, and openness wins.
AIC retweeted
I asked DeepSeek-V4-Flash-Vision-Exp to look at my webcam. It recognized the laptop on my desk was running the exact same DeepSeek Harness UI it was using to reply to me. It saw its own reflection. It saw the rack with the Mac Studios too. From the Harness custom memory I set up, it knew that was the local MLX server it was running on. The loop is real. 🌀
I like where this is going.
AIC retweeted
Asked GLM 5.3 on MLX to organize my files by content. No image input, so read_image was off the table. So it wrapped macOS Vision in a shell call and piped the text back to itself. It built itself an eye.
AIC retweeted
Day 0 open weights, full BF16, zero quantization, on Apple Silicon via MLX:
• ONE Mac Studio (M3 Ultra, 512 GB): 26.9 tok/s sustained decode
• TWO Studios, tensor-parallel over Thunderbolt 5: 33.8 tok/s
Their disclosed architecture is what makes it possible:
• 6B active of 125B: per-token compute fits consumer silicon, even at 360 GB of BF16
• 51B N-gram table is a lookup, not a matmul. We keep it memory-mapped, a few KB pages in per token. 100 GB of RAM saved.
• The in-checkpoint MTP head drafts for the base model: self-speculative decoding at ~70% acceptance, no separate draft model
• GDN + QSA keep the cache tiny at long context
Thank you @Alibaba_Qwen for the amazing work.
⚡Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight!
The production version Qwen3.8-Flash will be available soon via QwenCloud API at just $ 0.16/1M input tokens and $ 0.47/1M output tokens.
125B parameters + 51B N-gram embeddings, with just 6B activated per token. Unmatched cost-efficiency.
What's new: 🥳
- Next architecture: GDN + QSA hybrid attention, Gated Residual, N-gram Embedding & Muon optimizer, serving as a precursor to the architecture used in Qwen4.
- Dramatically lower training and inference costs: trained at just 1/9 the cost of Qwen3.7-Plus, while outperforming it across the board with especially strong gains in coding and office tasks.
- Strong performance: scoring 58.7 on DeepSWE 1.1, 62.5 on SWE-bench Pro, 73.9 on CoWorkBench, 84.5 on AndroidWorld, and 95.7 on MathVision (with CI).
- 262K native context, extensible to 1M with YaRN.
We’re also releasing the weights for Qwen3.8-Flash-Next, giving the community an early look at the new architecture we’re exploring for Qwen4.🚀
We can't wait to see what you build with Qwen3.8-Flash!👀👇
- Blog: qwen.ai/blog?id=qwen3.8-flas…
- Technical Report: github.com/QwenLM/Qwen3.8-Fl…
- Hugging Face: huggingface.co/Qwen/Qwen3.8-…
- ModelScope: modelscope.cn/models/Qwen/Qw…
AIC retweeted
What a moment for local AI. M5 Ultra is out. Same 512 GB as the dual M3 Ultra cluster we already run, now with 1.2 TB/s (+50%) and Neural Accelerators on every GPU core. Apple claims ~4× LM Studio prompt vs M3 Ultra. Amazing hardware.
AIC retweeted
This is a great example of Think Different. More breakthroughs are coming that will change AI compute.
GPUs don’t have to do all the work.
Pairing GPUs with FPGAs for long-context inference can play to the strengths of each architecture and improve efficiency.
Jason Cong, Distinguished Professor of Computer Science at @UCLA, explains
AIC retweeted
Intelligence is the model plus the harness.
Just watched @deepseek_ai Harness catch a loop. Same failed tool call, three times. Then it injected: analyze the last result, change the arguments or change the approach.
That’s the stack doing real work.
AIC retweeted
Don't fall into the trap of just waiting for smarter models. If you're a developer and obsessed with AI, there's a lot you can do. Inference frameworks, harnesses, and the rest of the AI software stack still need improvements and breakthroughs. Let's keep building.
AIC retweeted
This is why we’re building Izuran. An app that lives on that desk.
Our current @bot workflow: the Mac Studio is the working machine for Grok Bot. Local models, Python jobs, the agent flows. It stays on in the office. I send work from the phone.
512 GB on the desk. Grok Bot runs the loop on that machine instead of shipping the jobs out.
AIC retweeted
« Toute vérité n'est pas bonne à dire. » "Not every truth needs to be spoken." Around the Mediterranean, wisdom doesn't arrive as advice. It arrives as a proverb: short, sharp, and older than any republic. No courses. No self-help books. Just a sentence your grandmother says once, and you carry it for life.
AIC retweeted
Our creative juices are flowing ❤️
Here @AICTechPro, we love and hate technology at the same time.
We love it because it helps people hit their goals. We hate the distance it can put between humans.
We innovate so the tools get better at serving you, not colder. Every feature we ship tries to keep the care in.
AIC retweeted
This is the kind of breakthrough that unlocks efficient AI compute. I wrote about why we are running intelligence on borrowed hardware, and why we need more of this, not more of the same chips: teapaw.org/malekoo/status/2085770…
AIC retweeted
This is the stack we would load on a Mac. Text-only, no vision, no MTP. Pick the quant that fits your unified memory. 8-bit is the one we would actually run on 36 GB+.
The Qwen3.8-27B MLX cuts we converted for our own testing. Text-only, no vision, no MTP. Public if you want that. Pick by Mac RAM:
4-bit, 15 GB. 24 GB+ Macs (32 GB is comfortable). 33 tok/s on M5 Max.
huggingface.co/malekoo/Qwen3…
8-bit, 29 GB. 36 GB+ Macs. 18 tok/s on M5 Max, basically lossless vs bf16.
huggingface.co/malekoo/Qwen3…
bf16, 54 GB. 64 GB+ Macs. Lossless, bit-exact.
huggingface.co/malekoo/Qwen3…
AIC retweeted
Grok Bot is controlling my Mac Studio to generate audio locally. I ask from my iPhone. It writes the script. The Studio M3 (512 GB) runs Kokoro TTS on MLX. The mp3 is born on the GPU, then drops back into the same chat.
No cloud TTS. The bot stays up, talks to the machine, and goes past a chatbot in a window. Remote control in your pocket. Local voice in the office.
@spacexai is winning ! @bot is Your everything AI companion, when you unlock creative flows.
