Загрузка видео...

Не удалось загрузить видео

На главную

Whisper Large v3 Turbo running 100% local on Mac - powered by Apple Neural Engine! 🔥 Brought to you by argmax ⚡

reach_vb's profile picture

Vaibhav (VB) Srivastav

33,414 subscribers

51,389 просмотров • 1 год назад •via X (Twitter)

Наука и технологииМузыкаОбразование
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 10

Фото профиля Vaibhav (VB) Srivastav
Vaibhav (VB) Srivastav1 год назад

Weights on the Hub!

Фото профиля Vaibhav (VB) Srivastav
Vaibhav (VB) Srivastav1 год назад

Codebase:

Фото профиля Carlos Rene | DEGA.org
Carlos Rene | DEGA.org1 год назад

@argmaxinc Has anyone tried it on iOS for mobile devices?

Фото профиля Vaibhav (VB) Srivastav
Vaibhav (VB) Srivastav1 год назад

@argmaxinc Yes it works on iPhone (try the link in the GH repo)

Фото профиля Micah Engle-Eshleman
Micah Engle-Eshleman1 год назад

@argmaxinc Very cool! How long to transcribe an hour of audio?

Фото профиля caicaivic
caicaivic1 год назад

@argmaxinc I’ve been downloading for several hours, but the download hasn’t progressed at all. This issue happens with all models, and I don’t know what the problem is.

Фото профиля argmax
argmax1 год назад

@reach_vb Our open-source model files are hosted on Hugging Face Hub and it is generally pretty fast and reliable for downloads. Happy to help troubleshoot on our Discord:

Фото профиля Nabil Hunt 🌑 𝕏
Nabil Hunt 🌑 𝕏1 год назад

@argmaxinc Btw I was comparing whisper from OpenAi and "incredibly fast" one on replicate, and I noticed that consistently the timestamps for words are shifted and less accurate than the former ... Idk why

Фото профиля Furkan Gözükara
Furkan Gözükara1 год назад

@argmaxinc Still don't buy Mac if you don't want to pull your hair to make ai apps run

Фото профиля Rishabh Kedia
Rishabh Kedia1 год назад

@argmaxinc Hi Vaibhav, what is the value proposition? This maybe a very naive question but I’m curious to understand why will I use this library than directly use OpenAI realtime api or whisper?

Похожие видео

Local voice AI with a 235 billion parameter LLM. ✅ - smart-turn v2 - MLX Whisper (large-v3-turbo-q4) - Qwen3-235B-A22B-Instruct-2507-3bit-DWQ - Kokoro All models running local on an M4 mac. Max RAM usage ~110GB. Voice-to-voice latency is ~950ms. There are a couple of relatively easy ways to carve another ~100ms off that number. But it's not a bad start!
3:24

Local voice AI with a 235 billion parameter LLM. ✅ - smart-turn v2 - MLX Whisper (large-v3-turbo-q4) - Qwen3-235B-A22B-Instruct-2507-3bit-DWQ - Kokoro All models running local on an M4 mac. Max RAM usage ~110GB. Voice-to-voice latency is ~950ms. There are a couple of relatively easy ways to carve another ~100ms off that number. But it's not a bad start!

kwindla

66,639 просмотров • 11 месяцев назад

For the first time, the latest LLMs run on the Apple Neural Engine — and NexaSDK is the only framework that makes it possible, powered by the NexaML engine. Last year, our two co-founders were invited by Apple DMLI team (Data & Machine Learning Innovation) to share their research about on-device multimodal model for local AI agents. One of the big questions in the room was: “Can the newest LLMs actually run on ANE?” At the time, nobody had a clear path. Today, that path exists. NexaSDK now runs Granite-4.0 (IBM), Qwen3 (Qwen), Gemma3 (Google), and Parakeet-v3 (NVIDIA) fully on Apple’s NPU — unlocking low-power, always-on, fast inference across Mac and iPhone. A new wave of NPU-first local AI apps is coming to Apple devices. Start with one line of code on Mac. iOS SDK coming soon.
1:42

For the first time, the latest LLMs run on the Apple Neural Engine — and NexaSDK is the only framework that makes it possible, powered by the NexaML engine. Last year, our two co-founders were invited by Apple DMLI team (Data & Machine Learning Innovation) to share their research about on-device multimodal model for local AI agents. One of the big questions in the room was: “Can the newest LLMs actually run on ANE?” At the time, nobody had a clear path. Today, that path exists. NexaSDK now runs Granite-4.0 (IBM), Qwen3 (Qwen), Gemma3 (Google), and Parakeet-v3 (NVIDIA) fully on Apple’s NPU — unlocking low-power, always-on, fast inference across Mac and iPhone. A new wave of NPU-first local AI apps is coming to Apple devices. Start with one line of code on Mac. iOS SDK coming soon.

NEXA AI

30,213 просмотров • 8 месяцев назад

Fuck yeah! Llama 3.2 3B running on your browser! 100% local, powered by WebGPU & MLC 🦙
1:22

Fuck yeah! Llama 3.2 3B running on your browser! 100% local, powered by WebGPU & MLC 🦙

Vaibhav (VB) Srivastav

282,006 просмотров • 1 год назад

Gemma 2 2B running in a browser, powered by WebLLM & WebGPU! 🔥 100% local & on-device In less than 24 hours, we've already got the model to the edge! ⚡ Try it out on an HF space below:
1:31

Gemma 2 2B running in a browser, powered by WebLLM & WebGPU! 🔥 100% local & on-device In less than 24 hours, we've already got the model to the edge! ⚡ Try it out on an HF space below:

Vaibhav (VB) Srivastav

53,790 просмотров • 1 год назад

Introducing Shaders v3 🔥 Our biggest release yet. Built for design engineers who ship magic. Design production-ready WebGPU shader effects on an infinite canvas, powered by a new rendering engine that's up to 25× faster.
0:52

Introducing Shaders v3 🔥 Our biggest release yet. Built for design engineers who ship magic. Design production-ready WebGPU shader effects on an infinite canvas, powered by a new rendering engine that's up to 25× faster.

shaders

65,597 просмотров • 4 дней назад

Running Phi 4 Mini Instruct at 75 tok/sec locally on a Mac - powered by llama.cpp! 🔥 All you need is: brew install llama.cpp Followed by: llama-cli -hf bartowski/microsoft_Phi-4-mini-instruct-GGUF:Q8_0 That's it! 🤗
1:01

Running Phi 4 Mini Instruct at 75 tok/sec locally on a Mac - powered by llama.cpp! 🔥 All you need is: brew install llama.cpp Followed by: llama-cli -hf bartowski/microsoft_Phi-4-mini-instruct-GGUF:Q8_0 That's it! 🤗

Vaibhav (VB) Srivastav

30,964 просмотров • 1 год назад

Goodwood was born to be raced, especially by a Porsche 963. This car is powered by a 4.6-litre twin-turbo V8 engine with an electric motor, reaching speeds of up to 200mph! 😏
1:10

Goodwood was born to be raced, especially by a Porsche 963. This car is powered by a 4.6-litre twin-turbo V8 engine with an electric motor, reaching speeds of up to 200mph! 😏

RENNSPORT

72,237 просмотров • 3 лет назад

Run Qwen-TTS on Your Mac, natively. Blaine Brown has created Qwen3-TTS MLX WebUI Enhanced, a Web UI for running Qwen-TTS using the Mac-native MLX framework, powered by MLX-Audio from Prince Canuma, it runs very smooth! 1-click install on pinokio (Apple Silicon Macs)
0:40

Run Qwen-TTS on Your Mac, natively. Blaine Brown has created Qwen3-TTS MLX WebUI Enhanced, a Web UI for running Qwen-TTS using the Mac-native MLX framework, powered by MLX-Audio from Prince Canuma, it runs very smooth! 1-click install on pinokio (Apple Silicon Macs)

cocktail peanut

168,965 просмотров • 5 месяцев назад

Brought to you by our friends in Optics and Design. Watch #Severance Season 2 on Apple TV+
0:52

Brought to you by our friends in Optics and Design. Watch #Severance Season 2 on Apple TV+

Apple TV

215,672 просмотров • 11 месяцев назад

Qwen-2.5 on WebGPU 🏎️ • 42 tok/sec for Qwen2.5-Coder-1.5B on Mac ⚡ • Powered by MLC WebLLM and WebGPU 🔥 Watch Qwen2.5-Coder-1.5B build a website entirely in the browser!
0:35

Qwen-2.5 on WebGPU 🏎️ • 42 tok/sec for Qwen2.5-Coder-1.5B on Mac ⚡ • Powered by MLC WebLLM and WebGPU 🔥 Watch Qwen2.5-Coder-1.5B build a website entirely in the browser!

Caleb

41,579 просмотров • 1 год назад

Fuck it, 685B parameter, DeepSeek V3 0324 running locally on M3 Ultra, fully private 🔥 Powered by llama.cpp & dynamic quants from Unsloth AI ⚡ Step 1: brew install llama.cpp Step 2: llama-cli -hf unsloth/DeepSeek-V3-0324-GGUF:Q2_K_XL That's it! 🤗 Honestly a bit surreal to be able to chat with such a chunky model at the touch of the keyboard - future is going to be wild!!
5:00

Fuck it, 685B parameter, DeepSeek V3 0324 running locally on M3 Ultra, fully private 🔥 Powered by llama.cpp & dynamic quants from Unsloth AI ⚡ Step 1: brew install llama.cpp Step 2: llama-cli -hf unsloth/DeepSeek-V3-0324-GGUF:Q2_K_XL That's it! 🤗 Honestly a bit surreal to be able to chat with such a chunky model at the touch of the keyboard - future is going to be wild!!

Vaibhav (VB) Srivastav

168,141 просмотров • 1 год назад

Puli to Alexis for a sensational start ⚡ Brought to you by SNAIFUN
1:04

Puli to Alexis for a sensational start ⚡ Brought to you by SNAIFUN

AC Milan

59,028 просмотров • 9 месяцев назад

⚡️ NeuralAI Presents: World Engine Teaser 🌏 THE GAME ENGINE REIMAGINED Generate full scale maps, scenes, and entire worlds in collaboration mode with the World Engine, the ultimate AI-powered game engine! Real-time Unleashed Crisp Visuals at AAA Quality Real-time Asset Generation AI Agent Integration and Tokenisation ...And much more! Powered by $NEURAL Ready to create your world? 🚀 🧑‍🚀
0:55

⚡️ NeuralAI Presents: World Engine Teaser 🌏 THE GAME ENGINE REIMAGINED Generate full scale maps, scenes, and entire worlds in collaboration mode with the World Engine, the ultimate AI-powered game engine! Real-time Unleashed Crisp Visuals at AAA Quality Real-time Asset Generation AI Agent Integration and Tokenisation ...And much more! Powered by $NEURAL Ready to create your world? 🚀 🧑‍🚀

NeuralAI

99,117 просмотров • 1 год назад

Bugatti’s W16 Mistral Limited to 99 units at €5 million each. Powered by a 1,578-hp quad-turbo W16 engine, it hits 260 mph and goes 0 to 60 mph in 2.5 seconds
0:43

Bugatti’s W16 Mistral Limited to 99 units at €5 million each. Powered by a 1,578-hp quad-turbo W16 engine, it hits 260 mph and goes 0 to 60 mph in 2.5 seconds

Massimo

21,713 просмотров • 1 год назад

Introducing Moonshine Web: real-time speech recognition running 100% locally in your browser! 🚀 Faster and more accurate than Whisper 🔒 Privacy-focused (no data leaves your device) ⚡️ WebGPU accelerated (w/ WASM fallback) 🔥 Powered by ONNX Runtime Web and Transformers.js
0:40

Introducing Moonshine Web: real-time speech recognition running 100% locally in your browser! 🚀 Faster and more accurate than Whisper 🔒 Privacy-focused (no data leaves your device) ⚡️ WebGPU accelerated (w/ WASM fallback) 🔥 Powered by ONNX Runtime Web and Transformers.js

Xenova

27,878 просмотров • 1 год назад

Earn creator fees on every transaction, paid directly to your wallet. Powered by the $NEURAL ecosystem. The AI Worlds Launchpad is coming soon.
0:54

Earn creator fees on every transaction, paid directly to your wallet. Powered by the $NEURAL ecosystem. The AI Worlds Launchpad is coming soon.

NeuralAI

18,122 просмотров • 6 месяцев назад

This is INSANE. Google just dropped a free AI voice dictation app for iOS and Mac. Every premium feature works without any subscription. 100% free, runs fully local, powered by Gemma 4.
0:26

This is INSANE. Google just dropped a free AI voice dictation app for iOS and Mac. Every premium feature works without any subscription. 100% free, runs fully local, powered by Gemma 4.

Shubham Saboo

365,843 просмотров • 1 месяц назад

Announcing SubStudio! Generate subtitles for any video in seconds with AI. 100% free & open source! Powered by Whisper on Together AI and FFmpeg via fluent-ffmpeg.
0:20

Announcing SubStudio! Generate subtitles for any video in seconds with AI. 100% free & open source! Powered by Whisper on Together AI and FFmpeg via fluent-ffmpeg.

Hassan

133,261 просмотров • 3 месяцев назад

Run StableDAW on your PC with 1 click (No GPU needed) A full AI music studio powered by Stable Audio 3: generate tracks from text, edit them on a timeline, master them, even DJ them --- 100% local. Created by GANTASMO No GPU required. All OS supported: Windows, Linux, and Mac.
1:35

Run StableDAW on your PC with 1 click (No GPU needed) A full AI music studio powered by Stable Audio 3: generate tracks from text, edit them on a timeline, master them, even DJ them --- 100% local. Created by GANTASMO No GPU required. All OS supported: Windows, Linux, and Mac.

cocktail peanut

16,638 просмотров • 1 месяц назад