#inference
14 stories taggedinference.
AI Agents Could Soon Tidy Up Their Own Memory Mid-Task
A new system from Apple ML Research lets an AI agent decide for itself when to compress its own working memory, stopping it from forgetting important details at the worst possible moment.
Your MacBook Could Be Running AI Workloads While You Sleep
Apple ML Research has built a system called Minions that pools the idle GPUs inside MacBooks and Mac desktops to run the kind of AI tasks that normally require expensive server farms.
Liquid AI's vision model gets a speed boost of up to 3.13 times with no change to what it says
A 280-million-parameter add-on lets LFM2.5-VL-3B generate text up to 3.13 times faster on Apple silicon, leaving output word-for-word identical.
Hugging Face Lets You Run Compressed AI Models on Your Mac With One Line of Code
Hugging Face has built GGUF support into its Transformers library, letting Apple Silicon Mac users load and run lightweight AI models at speeds matching dedicated local-AI apps, no special setup required.
Anthropic and OpenAI Are Quietly Shopping for Small Data Centers in Europe and the US
Both labs are hunting for 20-to-30 megawatt sites in the UK, the Nordics, and the US, a sharp turn from the gigawatt-scale megaprojects they announced just months ago.
Qualcomm Lands AWS Partnership to Build AI Data Centre Chips, Shares Jump 10%
Amazon just handed Qualcomm its second major cloud endorsement in months, and the market noticed.
NVIDIA's New Vera Rubin Chip Does 30 Times More AI Work Per Watt Than Its Predecessor
Fresh benchmark data puts NVIDIA's Vera Rubin NVL72 far ahead on energy efficiency for the complex, multi-step AI tasks that companies are building into their products at speed.
Nvidia's $20 Billion Groq Bet Goes Live: What It Means for AI Speed
Nvidia's Groq 3 LPX racks are now in full production and will be online later this year, promising AI responses more than four times faster than OpenAI's current top speed.
An AI chip startup just tripled its valuation in six months. Here's who's paying and why.
Etched has raised $700 million at a $21 billion valuation, led by a Wall Street trading giant that actually tested the hardware first. The numbers are striking. The speed of the rise is even more so.
Groq raises $350 million after ditching its own chips for Nvidia's
Once an Nvidia rival, the startup is now an Nvidia customer. Here is what happened and what it means for the cloud companies racing to run your AI.
The AI Data Center That Fits in a Box: Inside Runware's Sonic Inference Pod
A startup wants to replace giant, water-hungry data centers with small, portable computing units that can be shipped anywhere there is a power socket. Here is what it is building, and whether it holds up.
Three Harvard Dropouts Just Raised $300 Million to Build the AI Chip That Rivals Keep Calling Impossible
Etched hit a $10.3 billion valuation in seven months by betting that specialised AI chips could beat general-purpose ones on speed and cost. Here is what that means for anyone who pays to run AI.
Microsoft plugs AMD's newest chips into Azure to handle the AI compute crunch
Three new virtual machine types land in Microsoft's cloud, each built around AMD's latest processors and aimed at a different piece of the AI workload puzzle.
The Electricity Problem at the Heart of Modern AI
Every AI data centre runs on a fixed power budget. The company that squeezes the most useful work out of each watt wins. Here is what that means in plain English.