“Pacing” is a strange word for a company that has been shipping at full speed, and OpenAI has made it official with a policy post that slows parts of its own frontier training. The August 18 announcement pauses Astra training workloads that do not yet meet new security standards and adds mandatory monitoring to reinforcement … Read more

The biggest change in code hosting this month is not coming from GitHub. Cursor, fresh off its $60 billion acquisition by SpaceX, is rolling out Origin, its own code hosting service, in early beta to all paid plan users starting this week. The pitch is simple: your repositories, pull requests, review comments, and AI agents … Read more

A 17GB model file that beats a closed flagship on the vendor’s own benchmarks is a rare thing, and Qwen 3.8 27B pairs it with a default setting that will quietly waste your whole evening. Released August 15 under an Apache 2 license, the 27B vision-capable model beats its predecessor and even a closed flagship … Read more

Keeping the base model frozen while scaling post-training is the bet behind GLM-5.3, and the numbers say it paid off. Zhipu shipped the update on August 14 with the exact same foundation as GLM-5.2, and a month of post-training scaling produced the strongest open-weights coding model the lab has shipped, plus an emergent cybersecurity capability … Read more

Closing a $60 billion acquisition took SpaceX four months, and the deal announced as complete on August 14 redraws the map of AI coding. Cursor now gets access to what it calls the largest fleet of GPUs in the world, and the first visible fruit of the pairing, Grok 4.6, shipped a week before the … Read more

An 8-billion-parameter language model and a 2.4-billion-parameter audio renderer sit behind MiniMax’s new music release, which turns a creative concept and optional lyrics into a complete, produced song in a single pass: Music 3.0, out August 13 with open weights, composes, arranges, performs, and produces a full track up to five minutes long, with vocals … Read more

“This time, it’s official.” That is how SiliconFlow opened its launch note for DeepSeek-V4-Pro-0813, and the production release largely backs up the hype: DeepSeek’s flagship model is out of preview with dramatically stronger agent benchmarks, native OpenAI Responses API support, and a shift to peak and off-peak pricing that takes effect August 17. What DeepSeek … Read more

Google’s workhorse tier of models has a new default, and it arrives at half the price of the one it replaces: Gemini 3.7 Flash, released August 13, ships with large gains on coding, document-heavy knowledge work, and agent benchmarks over 3.6 Flash, which launched only three weeks earlier. What Google Announced On August 13, Google … Read more

“Encrypted” is doing a lot of work in how leading LLM providers describe reasoning traces: new research shows the blocks can be swapped across sessions, users, and models, letting attackers force a weaker model from the same provider to decrypt and output a stronger model’s chain-of-thought in plaintext. What Was Found On August 10, a … Read more

Efficient reasoning is becoming the battleground for open model releases, and Ant Group’s Ling-3.0-tiny is the latest entry: a hybrid reasoning MoE with 7.9B total parameters and only 1.3B activated per token, in BF16, FP8, and INT4 variants, built for strong agentic and reasoning performance at low inference cost. What Was Released On August 11, … Read more