Forest City SFZ Clarifies Incentive Framework for Fintech and AI-Enabled Financial Services Alongside Malaysia Digital Incentives

FreeToken Unlocks Frontier MoE Inference on Consumer Hardware via Dynamic Co-Execution

Researchers from UC Berkeley and MIT have developed FreeToken, an open-source inference engine that enhances the utility of Mixture-of-Experts models on consumer hardware.

3h ago ·infoq.com

Why Does a 125B AI Model Use Only 6B Parameters at a Time?

Why Does a 125B AI Model Use Only 6B Parameters at a Time? A new AI model launches. You open X and immediately see numbers like: 125B parameters 6B active parameters 262K context Open weights And if…

• Model has 125 billion parameters but activates only 6 billion at once
• Mixture of Experts (MoE) architecture divides parameters into experts
• Router selects relevant experts for each input token, optimizing resource use

2h ago ·dev.to