Giuseppe Zangari

@dom_gag_96 Can I be added?

310

make eu great again make italy great again

@antirez They consider models as a commodity, they have a massive AI platform for enterprises and they are targeting that market. For many enterprise applications, the one you can automate, their models are pretty good also considering the hw required.

Btw anybody with some Mistral insider info is understanding WTF the company is doing? They are under delivering so much that there must be a hard reason, either politics, terrible management, total lack of GPU compared to Chinese players, or what?

@ivanfioravanti I really hope you are right and that is true!

My dream of reopening Olivetti to produce chips and hardware in Italy is more real now 🤔

@dom_gag_96 🍕

55 members- still growing

@mkurman88 @antirez I do not think that creating new LLM, at least in the way we know them today, is the right path for EU. But there is a lot to build in infra (if EU people prefer EU companies), harness, inference and applications. Also around socio-organizational approaches there room to innovate

Imho, before, the only blockers were a lack of money to start and interest from companies in investing, so now dealing with legal issues isn't helpful for innovation and fundraising. There are a few companies that try their best, but either they existed before law enforcement and have grown enough to pay for lawyers, or they are too small to gain more traction. Neither is seen as super successful because they don't spend much money on LLMs, which most people think of as AI, but on typical business-oriented values. Training a model from scratch costs so much money that no one in Europe is willing to risk it, especially given that it can easily fail without an extremely experienced team across various areas related to language models.

I want to say a final thing about my Fable first reaction: I dedicated my life to programming and I'll use every innovation in the field, also to extract value and bring it to the local inference world, to Redis, and so forth. But:

@vu3dtu I am happy to change idea folks! Show me a 7/20€ ring with open protocol I’ll buy it

I have been wearing apple watch, fitbit air and the ring for last 2 weeks. There is not a lot of variance in the step, heart rate and sleep surprisingly. Although a lot of second order derived metrics are harder to estimate like calories burnt. I have also been testing slightly more expensive rings $20-30 and they are even better.

I loved the idea behind the Google Fitbit Air: an LLM wrapped around your health data, daily briefs, and a coach you can ask questions. But there app is really terrible, it's expensive $100 band plus $10/mo, and Google getting a constant stream of your heart rate, sleep, and other private data. Whoop is worse, with a subscription that runs up to $360 a year. So I bought a $7 generic Chinese smart ring from Temu, reverse engineered its BLE protocol, and built an app around it. Introducing PulseLoop: no subscription, open-source iOS app. Your health data stays on your phone, paired with an AI coach that reads your real ring data, draws charts, and remembers context. Free, bring your own API keys. Demo and code below.

@mkurman88 @antirez It’s not the limit. The limit is adoption of EU solutions, market traction that in the end will give those companies power to innovate.

@antirez In the EU, we have a so-called AI Act. End of the story

@antirez That is true, this means also that we must believe in our digital companies and infrastructure. Only by believing and adopting our solution we can regain back control, and narrative.

So, what happened to use lately? Especially in Europe. In the US there are those few unicorns but where is all the rest of the AI scene? We need to recover our industrial ethics and stop accepting a narration that see ourselves boiled.

@the2ndfloorguy This is interesting and useful. Teams should include this metric too!

i hooked my whoop to my work calendar to find which coworker gives me the most stress 🚨 thanks to fable, I reverse engineered whoop to pull per minute heart rate. nd matched spikes with cal events and attendees I now have a leaderboard and I think about it daily. few info masked for obvious reasons ;)

@andst7 Nessun dubbio sulle abilità di Elon e sulla magnificienza di space-X, generare poi valore per le persone che hanno contribuito fin da subito è stupendo. Bisogna tenere conto che una IPO così è un movimento tettonico le cui conseguenze di 3/4 ordine non si conoscono. Da osservare.

Come sapete, SpaceX si quota in borsa questa settimana. Quello che non viene raccontato è che grazie alla quotazione, centinaia di saldatori, operai e meccanici vedono il loro patrimonio sfiorare il milione di dollari. Ricordo le storie di alcuni di loro: immigrati, studenti, disoccupati. All’inizio SpaceX diceva per queste figure: non servono capacità particolari, ma tanta voglia di imparare e lavorare. E forse è proprio così. Bravo Elon, bravi loro.

@morganlinton I have it since Thursday, I thin it’s great for what it is: fitness band to track parameters 24/7 except when training. When I do aerobic training I need a display. For the rest, I have “health sync” that sync with apple health and I use bevel free.

I'm returning my Fitbit Air. It was nice trying it out, definitely some good things about it, but overall the Google Health app is just incredibly disappointing, and there are some key missing features that I just can't live without. So I'm sticking with the WHOOP MG. More details on my decision, and the data behind it in the first comment below for those who missed my post this morning.

@morganlinton I think for this kind of workout is way better to have a screen, so I would go for my Apple Watch. It’s a pity they do not synchronize themself.

Beautiful trail run this morning. Continuing to test the Fitbit Air vs the WHOOP. Pretty disappointed with the Fitbit Air for exercise tracking. Rather than being able to track how long I spend in each zone, it just shows that I guess my run was “vigorous” As someone who does racing and triathlons this would be a total dealbreaker for me since I really try to optimize how much time I spend in zones 3 and 4. But I’m assuming this is just because it’s early, or maybe I’m missing something?

Sometimes the search for peace leads you close.

On AI, craft, and why process may become the real product. nuoveidee.substack.com/p/anyone-can-c… #AI #agent #transformation #LLM

@mvanhorn @rauchg @JinjingLiang @slashlast30days Added in my bookmark, cool!

@rauchg @JinjingLiang 26k stars on @slashlast30days built primarily in Claude code! github.com/mvanhorn/last3…

Nature always finds a way through the concrete. 🐌 #Nature #Adaptation

When AI agents will enter their accountability era? 😂💶 “Sorry for the misdirection.” Meanwhile your token balance: 📉 #ClaudeCode #AgenticAI #BuildInPublic #GenAI

Gallup simulates opinions with AI. Not replacement. Depth. The real innovation? Knowing what to ask next. #AI news.gallup.com/opinion/method…

🚨 The world’s first open-source 100B medical LLM is here 🏥 Local inferencers have a Health model option to run at home. AntAngelMed (100B params, only 6.1B active) recently released: ✅ Tops open-source models on MedBench & HealthBench ✅ 200+ tokens/sec on H20 hardware ✅ 128K context length 💪 Strong in medical reasoning, safety & empathy 🔒 Runs locally (full privacy) ⚡ Only 6.1B active params (very efficient) 🧠 Fine-tunable for hospitals & research 🖥️Practical Deployment Options for AntAngelMed (100B Medical LLM) (all are only estimates) ✅ Best Balance → INT4 (~50 GB) on 2–4x GPUs (RTX 5090 / 4090) ✅ Max Quality → FP8 (~100 GB) on (DGX Spark, Mac Studio 128GB) ✅ Budget Option → INT4 (~50 GB) on 2x RTX 4090 + CPU offload (slower) ❌ Single RTX 5090 (32GB) → Not recommended (model too big) ❔ GGUF could bring down the size even more Built by Zhejiang Health + Ant Healthcare. A big jump for open & privacy-friendly medical AI.