Prince Canuma
@Prince_Canuma
1.3K Following 22.9K Followers
One if the best advices from Buffet👌🏽
@Prince_Canuma Warren Buffett: "Be fearful when others are greedy, and be greedy when others are fearful."
You know it’s about to blow up when influencers and grifters start flooding it, 2023 déjà vu…
Local AI is a huge problem space with tremendous potential!
Meet Stormy. You hire it like anyone else: in a meeting.
- Walk it through your business, live call
- Runs your front desk and back office ops
- Answers calls and texts
- Fully generative software which builds itself as u use it
Mini demo: Stormy on a rental shop's front desk:
Show more
🔥 New Sparrow feature: table query with field filtering
Extract only the columns you need from a table. Sparrow uses dots.ocr to detect the table structure, then maps and type checks just the fields your schema asks for.
Watch example:
Code:
Try:
Show more
@Prince_Canuma deployed LocalMLXAI contract address is 0x40256876639561C23599F028E58926e292Da8bA3
view token:
Vibe coding only exists due to the lack of expertise and taste of the operator 👀
Gonna age like fine milk:)
Local models are useless
Vibing a custom inference engine is easy 🤣
Step 1: Spend $2,000 on tokens
Step 2: Qwen3.8 goes 3% faster
Step 3: New model drops, engine is obsolete
Step 4: Repeat until bankrupt or acquired
Show more
MLX-VLM is powering or is the reference for all the top 3 picks…
MLX Poll is showing interesting data!
Keep voting:
We need models optimized for 64GB machines.
Hands free mode better than…, just say:
“Hey Nativ”
"Sorry, I didn't get that."
"Sorry, I didn't get that."
"Sorry, I didn't get that."
Anyway, here's Hey Nativ 👇
Thanks for having me! ❤️
We started the day with a *successful* live demo of local AI models (it even did realtime voice recognition) by @Prince_Canuma
Making Mac’s go brrr 🚀
Based
Your are making claims that extrapolate the actual impact of this.
1. There is no mega kernel equivalent for GPU
2. Its a best case with 6 tokens accepted from draft model
3. Only valid for low volume of request. Is that a real use case?
vLLM is cooking, yes, but not it’s not as easy as that
Show more
This and @Nativ_AI live transcription in meetings.
The comparison with the early days of recorded music is interesting. Technology has always changed creative work, but it doesn’t necessarily erase the people behind it. The bigger question is whether AI changes what creativity looks like, and who gets to make a living from it.
Show more
Pretty cool 🔥🙌🏽
1/ Today we're introducing Cua-S1-4B-0.2, the first multimodal decision model trained with RLOO on live computer-use tasks, using task-completion rewards.
Text and multimodal adapters are available under Apache-2.0:
Show more
That was blazing fast!🔥
Nemotron 3 Diarization is now on MLX-Audio
Nativ now knows and remembers who you are 👌🏽
You know those long naps where you wake up and forget who you are?
Nativ doesn't.😌
Your Mac can doze off mid-recording and wake up with the recording saved. Nativ still knows your name, your style, and where that one chat is.
v0.3.9. Smoother than your Sunday nap.
Try it here:
Show more
Coming to MLX-Audio 🚀
@lllucas let’s go!
When several people talk at once, a transcript can get messy fast.
Our new Nemotron 3 Diarization model tracks who spoke when, even when voices overlap. It handles up to eight speakers, has 100M parameters, and is now available on @huggingface 🤗
Show more
This view is 10/10
I am getting ready to talk about @promptmotion and @CloudflareDev Computer at this beautiful venue @lisbonai_!
Nervous and excited. Let’s gooo ✨
