Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

The latest MLX has a CUDA back-end! To get started: pip install "mlx[cuda]" With the same codebase you can develop locally, run your model on Apple silicon, or in the cloud on Nvidia GPUs. MLX is designed around Apple silicon - which has a unified memory architecture. It uses...

awnihannun's profile picture

Awni Hannun

44,946 subscribers

42,761 görüntüleme • 11 ay önce •via X (Twitter)

Bilim & TeknolojiEğitim
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Day 1 of 3 MLX Releases: Introducing MLX-Audio 🚀🔥 A text-to-speech (TTS) and Speech-to-Speech (STS) library built on Apple's MLX framework, providing efficient speech synthesis on Apple Silicon. Features ⚡️Fast inference on Apple Silicon (M series chips) 🤖Multiple language support 🗣️Voice customization options 🚀Quantization support for optimized performance Supported models: 🪶Kokoro - A multilingual TTS model with 82M params that supports various languages and voice styles. With more models coming soon. Get started: > pip install mlx-audio Please leave us a star and send a PR :)
4:07

Day 1 of 3 MLX Releases: Introducing MLX-Audio 🚀🔥 A text-to-speech (TTS) and Speech-to-Speech (STS) library built on Apple's MLX framework, providing efficient speech synthesis on Apple Silicon. Features ⚡️Fast inference on Apple Silicon (M series chips) 🤖Multiple language support 🗣️Voice customization options 🚀Quantization support for optimized performance Supported models: 🪶Kokoro - A multilingual TTS model with 82M params that supports various languages and voice styles. With more models coming soon. Get started: > pip install mlx-audio Please leave us a star and send a PR :)

Prince Canuma

123,480 görüntüleme • 1 yıl önce

The video is a Llama v1 7B model implemented in MLX and running on an M2 Ultra. More here: * Train a Transformer LM or fine-tune with LoRA * Text generation with Mistral * Image generation with Stable Diffusion * Speech recognition with Whisper
1:35

The video is a Llama v1 7B model implemented in MLX and running on an M2 Ultra. More here: * Train a Transformer LM or fine-tune with LoRA * Text generation with Mistral * Image generation with Stable Diffusion * Speech recognition with Whisper

Awni Hannun

66,565 görüntüleme • 2 yıl önce

Nvidia announces the new RTX Spark, a new platform powered by the NX1 CPU, and shows off Spark laptops running 007 First Light and Forza 6. The CPU has 20 ARM based cores and a Blackwell RTX GPU with 6144 CUDA Cores. This is the same core count as a 5070, but with 128GB of unified LPDDR5X RAM memory sitting in the same package as the CPU and GPU. The entire Nvidia software stack is available, particularly CUDA, vital for AI. Nvidia's new laptops will likely be ideal for running local LLMs be cause the unified memory means you can load models up to 120-180B parameters (quantized). These laptops are expected to ship later this year and could become strong competitors to high-end MacBooks and even Mac Studios for local AI workloads, thanks to CUDA support and unified memory. Price is unannounced.
2:24

Nvidia announces the new RTX Spark, a new platform powered by the NX1 CPU, and shows off Spark laptops running 007 First Light and Forza 6. The CPU has 20 ARM based cores and a Blackwell RTX GPU with 6144 CUDA Cores. This is the same core count as a 5070, but with 128GB of unified LPDDR5X RAM memory sitting in the same package as the CPU and GPU. The entire Nvidia software stack is available, particularly CUDA, vital for AI. Nvidia's new laptops will likely be ideal for running local LLMs be cause the unified memory means you can load models up to 120-180B parameters (quantized). These laptops are expected to ship later this year and could become strong competitors to high-end MacBooks and even Mac Studios for local AI workloads, thanks to CUDA support and unified memory. Price is unannounced.

Grummz

30,573 görüntüleme • 1 ay önce

You can now run SQLCoder with a GUI on Apple Silicon or any NVIDIA GPU-enabled device! On Apple Silicon, just run CMAKE_ARGS="-DLLAMA_METAL=on" pip install "sqlcoder[llama-cpp]" sqlcoder launch The Apple Silicon version is not super accurate, but works great for simple questions – and everything happens (fairly fast) on just your laptop! We will add support for more platforms soon. PRs are very welcome! Github:
0:50

You can now run SQLCoder with a GUI on Apple Silicon or any NVIDIA GPU-enabled device! On Apple Silicon, just run CMAKE_ARGS="-DLLAMA_METAL=on" pip install "sqlcoder[llama-cpp]" sqlcoder launch The Apple Silicon version is not super accurate, but works great for simple questions – and everything happens (fairly fast) on just your laptop! We will add support for more platforms soon. PRs are very welcome! Github:

Rishabh Srivastava

10,405 görüntüleme • 2 yıl önce

Imagine a future where you have truly personal and proactive assistants on the best personal AI devices ever - your MacBooks. With MLX, that future is already here. Apple Silicon makes local AI genuinely powerful. We hosted MLX India’s first community meet-up on May 3rd. And here’s what people had to say.
2:44

Imagine a future where you have truly personal and proactive assistants on the best personal AI devices ever - your MacBooks. With MLX, that future is already here. Apple Silicon makes local AI genuinely powerful. We hosted MLX India’s first community meet-up on May 3rd. And here’s what people had to say.

sabesh 📟

20,616 görüntüleme • 1 ay önce

Day 1 of 3 days of MLX: Introducing MLX-Audio-Swift SDK 🚀 A modular Swift SDK for voice agents and tasks on Apple Silicon built by Lucas Newman and yours truly. iOS, macOS, and visionOS developers can now build native apps with real-time, on-device audio intelligence: 🗣️ Text-to-Speech (TTS) 👂 Speech-to-Text (STT) 🔄 Speech-to-Speech (STS) 🎙️ Voice Activity Detection (VAD) and more. Only import the capabilities you need, nothing extra. Get started today and leave us a star ⭐️
4:29

Day 1 of 3 days of MLX: Introducing MLX-Audio-Swift SDK 🚀 A modular Swift SDK for voice agents and tasks on Apple Silicon built by Lucas Newman and yours truly. iOS, macOS, and visionOS developers can now build native apps with real-time, on-device audio intelligence: 🗣️ Text-to-Speech (TTS) 👂 Speech-to-Text (STT) 🔄 Speech-to-Speech (STS) 🎙️ Voice Activity Detection (VAD) and more. Only import the capabilities you need, nothing extra. Get started today and leave us a star ⭐️

Prince Canuma

160,927 görüntüleme • 4 ay önce

🔴 The Pain: Running local MLX models is incredibly fast and private. But let's be real - testing tool calling via terminal is clunky, and there's zero good UI for it. 🟢 The Fix: rapid-mlx share Just ONE command gives you a polished web chat + seamless tool calling (works beautifully with gemma-4-12b-qat). We are proud to be the ONLY MLX inference engine in the community shipping this. ⚡️ 👇 Try it now: brew install raullenchai/rapid-mlx/rapid-mlx
0:40

🔴 The Pain: Running local MLX models is incredibly fast and private. But let's be real - testing tool calling via terminal is clunky, and there's zero good UI for it. 🟢 The Fix: rapid-mlx share Just ONE command gives you a polished web chat + seamless tool calling (works beautifully with gemma-4-12b-qat). We are proud to be the ONLY MLX inference engine in the community shipping this. ⚡️ 👇 Try it now: brew install raullenchai/rapid-mlx/rapid-mlx

raullen

22,680 görüntüleme • 1 ay önce

Sam-Audio by AI at Meta is now on MLX-Audio 🔊 You can now isolate voices or sounds from any audio locally in your Mac. I added this simple UI to the MLX-Audio studio app to help you get started Note: - The release is coming soon for now please install from source. - Vision part was excluded for now. Let me know if you need it. - PR for UI:
1:38

Sam-Audio by AI at Meta is now on MLX-Audio 🔊 You can now isolate voices or sounds from any audio locally in your Mac. I added this simple UI to the MLX-Audio studio app to help you get started Note: - The release is coming soon for now please install from source. - Vision part was excluded for now. Let me know if you need it. - PR for UI:

Prince Canuma

19,027 görüntüleme • 6 ay önce

Next mlx-vlm release will ship with continuous batching support on the server 🚀 What's coming: → Continuous batching — new requests join the active batch immediately, no waiting. Mixed image + text batches supported → OpenAI-compatible API — field-for-field match with mlx-lm, reasoning/content split for thinking models, tag-aware streaming → Multi-turn tool calling — full tool use support across streaming and non-streaming, works with Gemma4 and other templates → Vision feature caching — cache image embeddings across turns. Gemma4: 228x speedup, Qwen3.5: 23x on cache hit All running locally on Apple Silicon. Check our this demo running 4 concurrent requests (mixed image + text) to gemma-4-26B-A4B-IT by Google Gemma in bf16 using Pi + MLX-VLM server on my M3 Ultra. One of the requests ingests a 8K resolution image!
0:52

Next mlx-vlm release will ship with continuous batching support on the server 🚀 What's coming: → Continuous batching — new requests join the active batch immediately, no waiting. Mixed image + text batches supported → OpenAI-compatible API — field-for-field match with mlx-lm, reasoning/content split for thinking models, tag-aware streaming → Multi-turn tool calling — full tool use support across streaming and non-streaming, works with Gemma4 and other templates → Vision feature caching — cache image embeddings across turns. Gemma4: 228x speedup, Qwen3.5: 23x on cache hit All running locally on Apple Silicon. Check our this demo running 4 concurrent requests (mixed image + text) to gemma-4-26B-A4B-IT by Google Gemma in bf16 using Pi + MLX-VLM server on my M3 Ultra. One of the requests ingests a 8K resolution image!

Prince Canuma

82,169 görüntüleme • 3 ay önce

Introducing Marvis-TTS 🔥🚀 A new local-first TTS model Lucas Newman and I built for efficiency, accessibility, and real-time performance right on consumer devices like Apple Silicon, iPhones, iPads, and more. Traditional TTS models often demand full text inputs or sacrifice real-time capabilities, Marvis flips the script. It streams audio chunks as text is processed, creating a truly conversational experience. No more awkward pauses or unnatural breaks—Marvis handles the entire text context intelligently to deliver coherent, expressive speech. Get started today: > pip install -U mlx-audio
3:57

Introducing Marvis-TTS 🔥🚀 A new local-first TTS model Lucas Newman and I built for efficiency, accessibility, and real-time performance right on consumer devices like Apple Silicon, iPhones, iPads, and more. Traditional TTS models often demand full text inputs or sacrifice real-time capabilities, Marvis flips the script. It streams audio chunks as text is processed, creating a truly conversational experience. No more awkward pauses or unnatural breaks—Marvis handles the entire text context intelligently to deliver coherent, expressive speech. Get started today: > pip install -U mlx-audio

Prince Canuma

81,345 görüntüleme • 10 ay önce

🗓️ Release Week Recap Big week. mlx-audio and mlx-vlm are now among some of the fastest-growing OSS projects. Here’s what we shipped last week. Gemma 4 on Apple Silicon Two awesome releases by our partner Google DeepMind & Google Gemma : > Gemma 4 12B — their new dense, unified multimodal model. It uses an encoder free audio path and simplified vision encoder. > Gemma 4 QAT — quantization-aware training checkpoints, optimized to run locally on consumer GPUs and edge devices, compressing the model while preserving the quality you expect from Gemma 4. On the audio 🎧 side we added support for 15+ new TTS, ASR & VAD models, faster long-form transcription, and an expanded OpenAI-compatible audio server. All local on Apple Silicon. Huge thanks to every contributor and my co-maintainer Lucas Newman. 🙏🏽
1:34

🗓️ Release Week Recap Big week. mlx-audio and mlx-vlm are now among some of the fastest-growing OSS projects. Here’s what we shipped last week. Gemma 4 on Apple Silicon Two awesome releases by our partner Google DeepMind & Google Gemma : > Gemma 4 12B — their new dense, unified multimodal model. It uses an encoder free audio path and simplified vision encoder. > Gemma 4 QAT — quantization-aware training checkpoints, optimized to run locally on consumer GPUs and edge devices, compressing the model while preserving the quality you expect from Gemma 4. On the audio 🎧 side we added support for 15+ new TTS, ASR & VAD models, faster long-form transcription, and an expanded OpenAI-compatible audio server. All local on Apple Silicon. Huge thanks to every contributor and my co-maintainer Lucas Newman. 🙏🏽

Prince Canuma

11,731 görüntüleme • 1 ay önce

SAM-2 is apache license - you can indeed just build crazy apps with them without fear. This weekend I ported SAM-2 models to run optimally on Apple Silicon. I have also quantized all the SAM models to 16bit, 8bit, and 4bit variants (collection below) Repo is open source and pip installable. Comes with all you'll need for interactive video segmentation - - bidirectional propagation, - positive and negative clicks, - bounding boxes, - streaming masks back to user, - lightweight fast-api server that you can extend I basically asked Codex to port over the original Facebook sam2 codebase to mlx and optimize it for mlx. And it just worked. This demo for example is a 4bit base-plus model (~100MB download) To speed up, you can also do spatial or temporal downsampling. That speeds inference on my old macbook by ~3x and the hit in parity is negligible.
0:53

SAM-2 is apache license - you can indeed just build crazy apps with them without fear. This weekend I ported SAM-2 models to run optimally on Apple Silicon. I have also quantized all the SAM models to 16bit, 8bit, and 4bit variants (collection below) Repo is open source and pip installable. Comes with all you'll need for interactive video segmentation - - bidirectional propagation, - positive and negative clicks, - bounding boxes, - streaming masks back to user, - lightweight fast-api server that you can extend I basically asked Codex to port over the original Facebook sam2 codebase to mlx and optimize it for mlx. And it just worked. This demo for example is a 4bit base-plus model (~100MB download) To speed up, you can also do spatial or temporal downsampling. That speeds inference on my old macbook by ~3x and the hit in parity is negligible.

AVB

17,346 görüntüleme • 1 ay önce

just built a local AI waifu on my two macbooks by using the oss parallax inference library!!! not super smooth yet but parallax is still the easiest way to run llms locally on top of mlx - and no bigco can steal my waifu. still need folks to help her see/hear/draw since mlx has no legit wrapper yet. i’m running a campaign giving away mac minis + dgx spark. please please share your localhost ai model setup by using parallax (you don't need a gpu to try self-hosting you can just start with an extra macbook you borrowed from your friend like i did) or any fun app built on top of yourself host model to link below. come build sovereign intelligence with me 🫶
7:41

just built a local AI waifu on my two macbooks by using the oss parallax inference library!!! not super smooth yet but parallax is still the easiest way to run llms locally on top of mlx - and no bigco can steal my waifu. still need folks to help her see/hear/draw since mlx has no legit wrapper yet. i’m running a campaign giving away mac minis + dgx spark. please please share your localhost ai model setup by using parallax (you don't need a gpu to try self-hosting you can just start with an extra macbook you borrowed from your friend like i did) or any fun app built on top of yourself host model to link below. come build sovereign intelligence with me 🫶

Asuka Zheng🎀@ICML

178,998 görüntüleme • 7 ay önce

GLM-5 runs with mlx-lm on a single 512GB M3 Ultra in Q4. It's quite good in my initial testing and pretty fast as well. It generated a highly functional space invaders game using 7.1k tokens at 15.4 tok/s and 419GB memory. Thanks to Gökdeniz Gülmez and Tarjei Mandt for the port.
1:11

GLM-5 runs with mlx-lm on a single 512GB M3 Ultra in Q4. It's quite good in my initial testing and pretty fast as well. It generated a highly functional space invaders game using 7.1k tokens at 15.4 tok/s and 419GB memory. Thanks to Gökdeniz Gülmez and Tarjei Mandt for the port.

Awni Hannun

60,599 görüntüleme • 5 ay önce

oMLX is working really well as single machine inference engine for coding agents! Caching is managed perfectly (it can use a ton of disk space, be aware!) and oQ quantization delivers great results. Behind the scenes it uses the standard MLX building blocks (75% created by Prince Canuma 🙏): - mlx-lm - mlx-vlm - mlx-embeddings - mlx-audio I tested Qwen3.6-35B-A3B-oQ6 on M5 Max with two pi instances and it was fast and furious and leveraging cache like crazy as you can see in the video. Let me try to create some oQ versions (2,4,6?) of MiniMax M2.7 now and then I'll pass to distributed inference. I must win! 💪
3:57

oMLX is working really well as single machine inference engine for coding agents! Caching is managed perfectly (it can use a ton of disk space, be aware!) and oQ quantization delivers great results. Behind the scenes it uses the standard MLX building blocks (75% created by Prince Canuma 🙏): - mlx-lm - mlx-vlm - mlx-embeddings - mlx-audio I tested Qwen3.6-35B-A3B-oQ6 on M5 Max with two pi instances and it was fast and furious and leveraging cache like crazy as you can see in the video. Let me try to create some oQ versions (2,4,6?) of MiniMax M2.7 now and then I'll pass to distributed inference. I must win! 💪

Ivan Fioravanti ᯅ

19,384 görüntüleme • 2 ay önce

Congratulations to the Cohere team on releasing North Mini Code 1.0 🔥 We provide day-0 support on MLX thanks to our close collaboration with the Cohere team. North Mini Code 1.0 is a 30B param MoE model with 3B active. It runs at ~66 tok/s in BF16, truly impressive speeds before any compression. Here is a quick demo I built using Pi by Mario Zechner & Armin Ronacher ⇌ as my agent hardness. Get started today: 1. Install mlx-vlm from source: > uv pip install git+ 2. Start the server and power your favourite agents (pi, opencode, hermes and etc) > uv run mlx_vlm.server --model CohereLabs/North-Mini-Code-1.0 New release tomorrow
4:34

Congratulations to the Cohere team on releasing North Mini Code 1.0 🔥 We provide day-0 support on MLX thanks to our close collaboration with the Cohere team. North Mini Code 1.0 is a 30B param MoE model with 3B active. It runs at ~66 tok/s in BF16, truly impressive speeds before any compression. Here is a quick demo I built using Pi by Mario Zechner & Armin Ronacher ⇌ as my agent hardness. Get started today: 1. Install mlx-vlm from source: > uv pip install git+ 2. Start the server and power your favourite agents (pi, opencode, hermes and etc) > uv run mlx_vlm.server --model CohereLabs/North-Mini-Code-1.0 New release tomorrow

Prince Canuma

35,367 görüntüleme • 1 ay önce

QVAC SDK 0.12.0 is now live, bringing longer context, increased memory optimisation, new modalities, and broader ecosystem support directly to your device. Key Features and Updates: - TurboQuant KV-Cache Quantization: Fit much longer context in the same memory. TurboQuant, an algorithm from Google Research, compresses the KV cache by up to 5x, near-lossless. - Text-to-Video: Generate video from a text prompt, fully local, with the new wan2.1 model in the Diffusion addon - Apple Metal Performance for Flux2-klein: Diffusion on Apple Silicon now matches MLX performance, the native benchmark for Apple GPUs - Robot Control (new VLA addon): A GGML-based Vision-Language-Action addon brings fast, efficient robot control to edge devices - Coding Assistant / Harness Support: QVAC now works with OpenCode and OpenClaw as a local provider. A new @qvac/ai-sdk-provider package automates model registry and provider integration - Cross-Platform Voice: Text-to-speech and Parakeet transcription moved from ONNX to the GGML engine for better CPU and GPU support on macOS, iOS, Windows, Linux, and Android. Parakeet also adds long-term streaming diarization (tracking who spoke when on live audio) - Faster Lightweight Visual Classification: A new GGML-based Classification addon delivers millisecond-level classification, useful where a vision-language model (VLM) would be unnecessarily slow - Under the Hood: Fabric synced to llama.cpp v8828 (from v8189), plus GPU acceleration added to image-upscale models for faster results Full release notes:
1:29

QVAC SDK 0.12.0 is now live, bringing longer context, increased memory optimisation, new modalities, and broader ecosystem support directly to your device. Key Features and Updates: - TurboQuant KV-Cache Quantization: Fit much longer context in the same memory. TurboQuant, an algorithm from Google Research, compresses the KV cache by up to 5x, near-lossless. - Text-to-Video: Generate video from a text prompt, fully local, with the new wan2.1 model in the Diffusion addon - Apple Metal Performance for Flux2-klein: Diffusion on Apple Silicon now matches MLX performance, the native benchmark for Apple GPUs - Robot Control (new VLA addon): A GGML-based Vision-Language-Action addon brings fast, efficient robot control to edge devices - Coding Assistant / Harness Support: QVAC now works with OpenCode and OpenClaw as a local provider. A new @qvac/ai-sdk-provider package automates model registry and provider integration - Cross-Platform Voice: Text-to-speech and Parakeet transcription moved from ONNX to the GGML engine for better CPU and GPU support on macOS, iOS, Windows, Linux, and Android. Parakeet also adds long-term streaming diarization (tracking who spoke when on live audio) - Faster Lightweight Visual Classification: A new GGML-based Classification addon delivers millisecond-level classification, useful where a vision-language model (VLM) would be unnecessarily slow - Under the Hood: Fabric synced to llama.cpp v8828 (from v8189), plus GPU acceleration added to image-upscale models for faster results Full release notes:

QVAC

9,932,369 görüntüleme • 1 ay önce

I trained a 12M parameter LLM on my own ML framework using a Rust backend and CUDA kernels for flash attention, AdamW, and more. Wrote the full transformer architecture, and BPE tokenizer from scratch. The framework features: - Custom CUDA kernels (Flash Attention, fused LayerNorm, fused GELU) for 3x increased throughput - Automatic WebGPU fallback for non-NVIDIA devices - TypeScript API with Rust compute backend - One npm install to get started, prebuilt binaries for every platform Try out the model for yourself: Built with Reese Chong. Check out the repos and blog if you want to learn more. Shoutout to Modal for the compute credits allowing me to train on 2 A100 GPUs without going broke cc sunny madra Gavin
2:12

I trained a 12M parameter LLM on my own ML framework using a Rust backend and CUDA kernels for flash attention, AdamW, and more. Wrote the full transformer architecture, and BPE tokenizer from scratch. The framework features: - Custom CUDA kernels (Flash Attention, fused LayerNorm, fused GELU) for 3x increased throughput - Automatic WebGPU fallback for non-NVIDIA devices - TypeScript API with Rust compute backend - One npm install to get started, prebuilt binaries for every platform Try out the model for yourself: Built with Reese Chong. Check out the repos and blog if you want to learn more. Shoutout to Modal for the compute credits allowing me to train on 2 A100 GPUs without going broke cc sunny madra Gavin

Aadi Kulshrestha

812,585 görüntüleme • 3 ay önce

PowerInfer - a high-speed inference engine for deploying LLMs locally. Just came across this super interesting project on speeding up inference. It's not MoE but it's a simple approach that exploits the high locality in LLM inference to design a GPU-CPU hybrid inference engine. Hot-activated neurons are preloaded onto the GPU for fast access, while cold-activated neurons (the majority) are computed on the CPU. This approach significantly reduces GPU memory demands and CPU-GPU data transfer. It achieves an average token generation rate of 13.20 tokens/s, with a peak of 29.08 tokens/s, across various LLMs on a single NVIDIA RTX 4090 GPU. It's on only 18% lower than that achieved by a top-tier server-grade A100 GPU. It also significantly outperforms llama.cpp by up to 11.69x while retaining model accuracy. There is a lot more innovation around inference that's coming fast. Really encouraged by the study on sparse computation to enhance the computational efficiency of LLMs. It's now possible to use PowerInfer with Llama 2 and Faclon 40B. Mistral-7B support is coming soon!
0:56

PowerInfer - a high-speed inference engine for deploying LLMs locally. Just came across this super interesting project on speeding up inference. It's not MoE but it's a simple approach that exploits the high locality in LLM inference to design a GPU-CPU hybrid inference engine. Hot-activated neurons are preloaded onto the GPU for fast access, while cold-activated neurons (the majority) are computed on the CPU. This approach significantly reduces GPU memory demands and CPU-GPU data transfer. It achieves an average token generation rate of 13.20 tokens/s, with a peak of 29.08 tokens/s, across various LLMs on a single NVIDIA RTX 4090 GPU. It's on only 18% lower than that achieved by a top-tier server-grade A100 GPU. It also significantly outperforms llama.cpp by up to 11.69x while retaining model accuracy. There is a lot more innovation around inference that's coming fast. Really encouraged by the study on sparse computation to enhance the computational efficiency of LLMs. It's now possible to use PowerInfer with Llama 2 and Faclon 40B. Mistral-7B support is coming soon!

elvis

261,622 görüntüleme • 2 yıl önce

"Mac Minis for example are a very good fit" - Andrej Karpathy Andrej Karpathy shouted out my work on EXO Labs in his keynote at Y Combinator AI SUS! Here's the breakdown: Right now most AI workloads run in the cloud where requests from different users are continuously batched together. These workloads are FLOPS-bound and favors hardware with the best unit economics of $ per FLOP, i.e. enterprise GPUs. The personal computing revolution will shift these workloads to personal devices with lower batch sizes (mostly batch_size=1). batch_size=1 inference is memory-bound, because all of the model parameters need to be loaded into the GPU every time a token is generated. Apple Silicon with its Unified Memory architecture has a lot of memory and memory bandwidth per $ compared to other hardware: - M4 Pro Mac Mini, 24GB @ 273GB/s, $58.33/GB, $5.13/GB/s - H100, 80GB @ 3350GB/s, $625/GB, $14.93/GB/s The unit economics of Apple Silicon are becoming more compelling with every release of the Mac. The future of AI inference looks more like open weights models (OpenAI open weights model soonTM) run at low batch_size on personal devices.
0:20

"Mac Minis for example are a very good fit" - Andrej Karpathy Andrej Karpathy shouted out my work on EXO Labs in his keynote at Y Combinator AI SUS! Here's the breakdown: Right now most AI workloads run in the cloud where requests from different users are continuously batched together. These workloads are FLOPS-bound and favors hardware with the best unit economics of $ per FLOP, i.e. enterprise GPUs. The personal computing revolution will shift these workloads to personal devices with lower batch sizes (mostly batch_size=1). batch_size=1 inference is memory-bound, because all of the model parameters need to be loaded into the GPU every time a token is generated. Apple Silicon with its Unified Memory architecture has a lot of memory and memory bandwidth per $ compared to other hardware: - M4 Pro Mac Mini, 24GB @ 273GB/s, $58.33/GB, $5.13/GB/s - H100, 80GB @ 3350GB/s, $625/GB, $14.93/GB/s The unit economics of Apple Silicon are becoming more compelling with every release of the Mac. The future of AI inference looks more like open weights models (OpenAI open weights model soonTM) run at low batch_size on personal devices.

Alex Cheema

103,487 görüntüleme • 1 yıl önce