DISPATCH // AI NEWS
AI and Prompt Engineering News, Updated Daily
Daily coverage of model releases, agent tooling, LLM inference and routing, and prompt-engineering technique, written for people shipping with these models.
IndustryAug 26, 20262 min read
OpenAI's Jalapeño Chip Beats Current Inference Hardware on Tokens and Efficiency
OpenAI's custom Jalapeño inference chip outperforms existing hardware on both throughput per user and tokens per kilowatt, according to SemiAnalysis benchmarks. For teams shipping LLM features at scale, this signals a meaningful shift in the cost and latency floor for OpenAI-hosted inference.
ModelsJul 28, 20262 min read
Kimi K3 Weights Drop: 2.8 Trillion Parameters, Open for Self-Hosting
Moonshot AI has released the full weights for Kimi K3, a 2.8 trillion parameter model, on Hugging Face. At 1.56TB, it is now the largest openly available model weights for teams willing to run their own inference.
ModelsJul 27, 20262 min read
Kimi K3 Goes Open-Weight: What Practitioners Need to Know
Moonshot AI's Kimi K3 is dropping open weights, giving local-deployment teams a new frontier model to fine-tune and self-host. Here is what changes for builders running their own inference.
AgentsJul 21, 20263 min read
Privacy Is Becoming the Default Design Constraint for AI Agents
Three independent signals on the same day point to a converging pattern: developers building AI agents are now treating privacy as a first-class architectural requirement, not an afterthought. Here is what that shift means for teams shipping agent features.
IndustryJun 25, 20262 min read
OpenAI Reveals Jalapeño: Its First Custom AI Inference Chip, Built with Broadcom
OpenAI and Broadcom have unveiled Jalapeño, a custom ASIC built specifically for LLM inference at data-center scale. For teams shipping LLM features, this signals a future where OpenAI controls more of its own inference stack, with direct implications for cost, latency, and API pricing.