@apirela_dev

low level engineering @massive_com | creating @fusor_rs | building cool stuff — ex @supabase

 Orlando, FL
Joined June 2020
  • Tweets1.3K
  • Following161
  • Followers591
81Media
1) i've been building something for my family. there's someone new at home.
1
3
2
12
446
Andres Pirela retweeted
Massive
@massive_com3h
Big data needs bigger infrastructure. We've 2x Massive's New York data center capacity, with room to scale. Shoutout to our platform engineering team for making it happen. 💙
1
3
142
I love how for Anthropic this was a simple release, for OpenAi it took hype, DevDay, vague posting, all to end up in the same place: being behind.
Vishal Singh
@vishalsingh297222h
WHAT THE ACTUAL FUCK?! Anthropic just decided to go after OpenAI Dots and Grok Bot with Claude Duet, powered by the absolute MONSTER that is Opus 5.5. We literally go to sleep, wake up, and somehow Anthropic has dropped ANOTHER product to take on its competitors. LIKE HOW ARE THESE GUYS SHIPPING THIS FAST?! THEY JUST CAN'T STOP WINNING. What an absolutely LEGENDARY run. 😭
1
162
Andres Pirela retweeted
backchannel
@heybackchannelOct 21
Two small open TTS models dropped this week: Ant Group's Ming-3.0-tts-0.5B and Gradium's Phonon-31M. Within a day, @apirela_dev had both running locally in the browser on CPU. He used fusor-ai, a Rust toolkit he built that compiles models into WebAssembly programs. His benchmarks on an M4 Pro: - Phonon — 2.4x faster than Gradium's browser CPU runtime with 8 threads, with a 48.5 MB download vs. 69.4 MB. - Ming — first audio in 0.3s vs. 14.3s for the non-streaming PyTorch reference, with quantized weights. He also shares where his implementation loses: Gradium's native CPU runtime is 3.5x faster on a single thread. "I'm all for privacy models running on CPU," he says. Both implementations are open source and still at v0.1. phonon-wasm.fusor.build ming-wasm.fusor.build #BackchannelSignals
Andres Pirela
@apirela_devOct 20
Ming-3.0-tts-0.5B now runs 100% locally in your browser. No server. No GPU. Just your CPU and WebAssembly. 5.5× faster than the PyTorch reference on the same machine, with the first audio in 0.3 s. Built in Rust with fusor-ai 🦀
2
2
11
751
Introducing Parakeet WASM: speech to text that runs entirely in your browser 116 s of audio -> 3 s. On your CPU. No server. No GPU. No API key. Your audio never leaves the tab. Moondream's Parakeet Redux, built with fusor-ai for WASM compatability. Try it now: parakeet-stt.fusor.build
3
1
1
59
it transcribes at real-time speed, and does not use GPU. It's great to embed in edge hardware, or frontends without having to pay a subscription. This was made with love by @fusor_rs for @moondreamai <3
19
If you like what I’m doing, follow @fusor_rs . That’s the real open source lab.
2
71
Hey @GradiumAI, I ended up running phonon in the browser using Fusor AI toolkit instead of xn-ptts. Here's a quick demo on hugging face: huggingface.co/spaces/fusorl… Here's the repo: github.com/fusor-rs/phonon-w… it runs 2.5x faster across 8 threads, and about 1.3x faster across 1 thread. It's very instant, the use cases for this are insane. Great job with Phonon, it feels instant already without any optimization. looking forward to doing more!

phonon-wasm - a Hugging Face Space by fusorlabs

Phonon 31M text to speech in the browser

huggingface.co
1
1
2
141
(on CPU)
16
Ming-3.0-tts-0.5B now runs 100% locally in your browser. No server. No GPU. Just your CPU and WebAssembly. 5.5× faster than the PyTorch reference on the same machine, with the first audio in 0.3 s. Built in Rust with fusor-ai 🦀
2
2
2
9
7,131
It's lighter too: 1.37 GB in a Chrome tab vs 6.1 GB for PyTorch. The model is an 850 MB download, cached after the first visit. After that, nothing you type leaves the page. Try it: huggingface.co/spaces/fusorl… Code: github.com/fusor-rs/ming-was…
1
6
93
Built by @fusor_rs for @AntLingAGI
3
59
I just compiled Ming 3.0 TTS (0.5B) to Web Assembly. It's really fast & runs fully in your browser, no API needed. I'll have more to share tomorrow! btw: ming 3.0 tts is nice!
Ant Ling
@AntLingAGIOct 19
Today, we're open-sourcing 5 Ming-3.0 models across 3 speech families. - Ming-3.0-tts: 0.5B voice design & paralinguistic cues; Tiny 7.9B A1.3B long-form speech. - Ming-3.0-audio: Flash 126B A6.5B understanding & tools; Tiny 9.3B A2.7B efficient dialogue. - Ming-3.0-asr: Tiny 8.6B A2B dialects & context.
1
66
A month ago I shipped fusor, a Rust web framework where the UI is plain HTML. It turned into more than one framework, so today it gets a name: Fusor Labs. We make tools. Then we use them. @fusor_rs
1
2
1
8
316
more replies
I'm particularly excited about Hyperpane, it lets you create native apps in Rust, HTML and can be used across your Mac, or iOS devices!
1
28
Two rules we hold ourselves to: using anything we make should take one CLI command, no tutorial. And things stay open source. fusor, HyperCMD and SQL Bomb are MIT today. It's early and the APIs still move. If something breaks, tell me. fusor.build/blog/introducing…

Introducing Fusor Labs · Fusor Blog

It all started as a framework for writing reactive sites in Rust with plain HTML. Now we want to push the limits further.

fusor.build
1
14
My mom proceeds to tell me she's been using Muse. I was surprised since she's the less techy person I know. we're definitely in a bubble
26
interested in using Rust and plain HTML for iOS apps? Follow @fusor_rs
49
if you want to meet Hughie, go sign up. he'll be able to order pizzas and whatnot
Andres Pirela
@apirela_devOct 16
1) i've been building something for my family. there's someone new at home.
2
62
In a turn of events & building in public, I'm happy to announce you can create iOS apps with Rust and plain HTML. they're also reactive by default. This should be coming sometime this week, follow along @fusor_rs
1
2
135
Load more