inference

AI

Laya runs offline on M4, ChatGPT tracks ads

New offline inference benchmarks, cross‑site data collection, and rescue services raise fresh questions for AI developers.

AI

NVIDIA unveils four AI inference platforms with Ada

NVIDIA launches four specialized inference platforms featuring the L4 and H100 NVL GPUs, aiming to speed generative AI workloads across cloud and enterprise.

AI

OpenAI halves GPT‑5.6 Sol price as Moonshot AI rolls out cheaper

OpenAI cuts GPT‑5.6 Sol fees by 50% while Moonshot AI launches Kimi K2.5 and K3, models that beat GPT‑5 on reasoning and bring trillion‑parameter open‑source to the market.

AI

Kog Boosts GPU Inference

Kog challenges the notion that GPUs are ill-suited for agentic workflows, aiming to squeeze more inference out of them. The French startup's approach may change how we utilize GPUs.

AI

LLM routers face churn as projects deprecate, refocus

Manifest shuts down its router while open‑source alternatives like RouteLLM, Any‑LLM, and ClawRouter vie for the cost‑saving niche.

AI

Groq Raises $650M

Groq shifts focus to AI inference, raising $650M in funding. What's next for the chipmaker?