Jan 31, 2026
They found bugs, debated consciousness, and founded a religion. Security researchers are alarmed.
Jan 30, 2026
Arcee AI released Trinity Large, which is a sparse MoE with 400 billion parameters, trained on 17 trillion tokens. Here's the complete technical breakdown.
Jan 29, 2026
Kimi K2.5: Understanding the Agent Swarm Architecture
Jan 27, 2026
How local AI inference works under the hood
MCP Apps shipped yesterday. The technical spec matters less than who signed off on it: direct competitors building shared infrastructure. Here's what actually happened.
Jan 26, 2026
A deep look at the open-source project that operates on outdated laptops and costs less than your streaming subscriptions
Jan 24, 2026
A Technical Breakdown of How AI Values Are Encoded
Jan 23, 2026
Understanding autonomous development loops
Jan 22, 2026
Liquid AI has released LFM2.5-1.2B-Thinking, a reasoning model designed for on-device deployment. The model contains 1.2 billion parameters and operates within 900 MB of memory on mobile devices.
Jan 20, 2026
Treating the GPU memory like an operating system