Valkey 9.1: Hybrid Search Kills the Two-Cluster Stack
Valkey 9.1 ships hybrid search combining full-text, vector, and numeric queries in one index—plus 10% memory savings and AI maintainer agents. What changes ...
Deno 2.9: deno desktop Ships, Cold Startup Hits 17ms
Deno 2.9 ships deno desktop for native cross-platform apps, cuts cold startup to 17ms, and ...
Amazon Bedrock AgentCore Harness Is GA: Two API Calls
Amazon Bedrock AgentCore Harness is now GA. Deploy production AI agents with CreateHarness and InvokeHarness ...
Fable 5 Ban Backfired: Asian AI Models Fill the Void
The US export ban on Fable 5 triggered Asian AI model launches this week. GLM-5.2, ...
DeepSeek DSpark Goes Live with 80% Inference Speed Gains
DeepSeek DSpark is live in production: 51-400% throughput, 80% latency reduction. DeepSpec open-sourced under MIT. ...
OpenAI Jalapeño Chip: 50% Cheaper Inference Explained
OpenAI and Broadcom unveiled Jalapeño, a custom LLM inference chip claiming 50% lower token costs than Nvidia GPUs. Here is what developers actually ...
Liquid AI LFM 2.5-230M: 230M Model Beats 1B Transformer on Edge
Liquid AI LFM 2.5-230M outperforms models 4x its size on data extraction and runs at ...
Agility Robotics IPO: $2.5B Humanoid SPAC Unpacked
Agility Robotics IPO at $2.5B via SPAC — the first humanoid robot public company. But ...
Google’s AI Brain Drain Just Cost Alphabet $270B
Four of Google’s most consequential AI researchers resigned in a single week. By the time ...
AI Data Centers Are Driving Up Your Electricity Bill
Voters are ousting politicians who approve AI data centers. Here's the electricity cost math, the ...