Full archive

Latest AI News

Search reporting on AI markets, models, products, companies, policy, and culture.

Showing 1-6 of 6Results matching "Apple Silicon"

Today

1 story

2 hours agoInfrastructure

Open release Infrastructure

Perplexity Open-Sources Lily, a Qwen-Specific Mac Engine That Beats MLX-LM in Its Test

The Rust-and-Metal runtime gives developers a public, deployable version of Perplexity’s local inference stack. Its reported gains come from one Qwen model, one hardware family and a batch-one comparison on a high-memory M5 Max.

3 min

Yesterday

1 story

14 hours agoStartups

Funding Startups

Lyte Raises $165M at $1.6B Valuation for Robot Perception

The Series C places a $1.6 billion price on Lyte’s combination of custom chips, sensing and AI software, with warehouses and factories as its initial market.

2 min

Earlier this week

1 story

YesterdayProducts

Launch Products

Perplexity Routes Sensitive Computer Tasks to Local Macs With a Privacy Gate

The cloud-first system can delegate confidential portions of an agent job to a local model without restarting it. Its privacy promise ultimately depends on a classifier and user decisions about what can leave the device.

3 min

Last week

2 stories

Aug 25, 2026Products

Launch Products

Apple Refreshes Mac mini and Mac Studio for Linked Local AI, From $899

The specification refresh formalizes a workflow of linking Macs for larger local models, but performance evidence has yet to arrive and the top chip’s name is inconsistent.

2 min

Aug 24, 2026Models

Open release Models

MiniMax Maps H3 to 24GB PCs and Server Stacks, but Keeps Context Layer Hosted

The index gives developers more routes to run and adapt H3, yet the layer that turns mixed media into model-ready context remains an API product—and local use is restricted in several major markets.

3 min

August 2026

1 story

Aug 17, 2026Tools

Research Tools

Braid’s Local AI Test Chooses Provenance Over Vector Portability

The demonstration shows a workable handoff between two named local embedding models, but it does not yet establish that meaning will hold across broader model families, payloads, or hostile conditions.

4 min