Day 1: Benchmarks We ran 1000+ LLM benchmarks on... real consumer devices. Data includes single-device and multi-device clusters with Tokens-Per-Second (TPS) and Time-To-First-Token (TTFT). Setups tested: 3x M4 Mac Mini cluster, iPhone 15 + S24, RTX4090 & more.show more

M4 Mac Mini AI Cluster Uses EXO Labs with... Thunderbolt 5 interconnect (80Gbps) to run LLMs distributed across 4 M4 Pro Mac Minis. The cluster is small (iPhone for reference). It’s running Nemotron 70B at 8 tok/sec and scales to Llama 405B (benchmarks soon).show more

M4 Mac Mini AI cluster running Llama 3.3 70B... Preliminary results show Gigabit ethernet switch (4 tok/sec) is slower than Thunderbolt-5 with EXO Labs Benchmarks including Thunderbolt-5 coming soonTM.show more

Distributed training on M4 Mac Mini cluster We implemented... Google DeepMind DiLoCo on Apple Silicon to train large models with 100-1000x less bandwidth compared to DDP baseline. AI is entering a new era where a distributed network of consumer devices can train large models.show more

m4 mac mini cluster with 5 nodes running exo

What if your cloud agents could use iOS and... Android devices just like that? Introducing Agent Device Cloud: hosted on your dusted Mac Mini, used through 𝚊𝚐𝚎𝚗𝚝-𝚍𝚎𝚟𝚒𝚌𝚎show more

Introducing Find the best local models based on real... data 1. People run and submit benchmarks 2. Stats aggregated over models / devices 3. Find the best model for you `npx whatcanirun`, fully open-sourceshow more

This is not possible with chatgpt… Only possible with... an on-device LLM: - Real-time note monitoring - Adjusts difficulty in real-time - Announces progressshow more

Today, we're shipping MLX support for TADA, our open-source... text-to-speech model, which means the entire pipeline (LLM, flow-matching, and decoder) can now run locally on any Apple Silicon device. We're seeing a 45% reduction in memory usage and a 10x speed-up when using it quantized. With these improvements, you can use TADA on-device for OpenClaw or any personal chatbot. If you own a MacBook, Mac Mini, or Mac Studio, record a 10-second clip of any voice, type any text, and get high-quality, natural and expressive speech in real-time. Completely offline, completely free.show more

Introducing Nomic Embed - the first fully open long... context text embedder to beat OpenAI - Open source, open weights, open data - Beats OpenAI text-embeding-3-small and Ada on short and long context benchmarks - Day 1 integrations with @langchain, Llama-index, MongoDBshow more

Multi-chain data, streamed in real-time. Get live prices, onchain... trades & OHLCV data for 20M+ tokens with the CoinGecko WebSocket API.show more

Mirror Xiaomi device on your iPhone, reply to messages,... and transfer files with ease! #XiaomiHyperOS3 #CrossEcosystemConnectivityshow more

No matter where you go, and no matter which... device, we believe your internet should go with you 🌏 Starting today, your Arc tabs, folders, & spaces SYNC in real-time across iPhone (Arc Search), Mac, & Windows. Android in the works. Here’s a demo from the new Arc Search app:show more

Just dropped on HF — NeuTTS Air Next-gen on-device... TTS that matches cloud-level quality while staying fully open source. > Real-time speech synthesis on CPU/GPU > 3-second voice cloning, no cloud or data upload > Compact: under 200 MB, runs on mobile and edge devices > Multilingual and expressive > Developed by Neuphonic , optimized for speed and fidelityshow more

The Expansion of LayerAI Network with laiOS 🧬 Each... laiOS device requires at least 1 token per device to operate. There are over 1 trillion devices in the world already. This number grows by tens of billions every year. If just 0.3% of devices opt into the network, that’s a demand for 3 billion $LAI. LayerAI aims to present the first working version of this product by Q1-Q2 of this year.show more

The newer iPad Pros run 8B models easily. Here’s... Llama running on-device at 30 tokens per second. Impressive for an iPad.show more

A startup in China packed roughly 1,000 Mac Mini... M4 computers into a single data center to run AI workloads without paying ongoing cloud fees. The base Mac Mini M4 starts at $599 and draws just 10 to 30 watts under load, a fraction of the 300 to 500 watts a traditional GPU server pulls.show more

Running multiple accounts on one device was never supposed... to work at scale. Yet professionals do it every day without bans, flags, or leaks. Octo Browser is the reason. A next-gen anti-detect infrastructure built to replicate real devices, secure data, and manage hundreds of accounts seamlessly. Multi-accounting isn’t the risk anymore; weak tools are. 👇show more

Wow, Apple finally catches up with Android here. Device... Hub lets you control a connected iPhone directly from your Mac without constantly reaching for the deviceshow more

✅ BENCHMARKED: TensorFold made Qwen3.8 27B more than 3x... faster on my 64GB M5 Pro Mac mini, on two versions in one day. In the morning I tested 1.2.1. On its own, the model wrote about 17 tokens a second. I then switched on the DFlash2 drafter, a small helper that guesses the next few tokens so the big model only has to check them and it jumped to 56. And later in the afternoon Ash Hart shipped 1.2.2. His release notes focused on other hardware, so I reran everything on my Mac mini. On mine it runs within a few percent of 1.2.1, and the drafter still makes it 3x faster. With randomness off, the answer came out word for word the same as without the drafter. - Chat, 3.26x faster on 1.2.1 and 3.35x on 1.2.2 - A 31K token prompt, 2.94x and 2.77x - Code, 5.4x and 5.5x, up to 91.90 tokens a second - 28.5 GiB peak memory on both - New in 1.2.2, the first request doesn’t wait for the drafter to load A massive thanks to Ash Hart and all the contributors behind 1.2.2. The TensorFold challenge from Yukon is open until Nov 6th if you want to try pushing it further on your own Mac.show more