Forums
Showing 12 discussions
Hugging Face says open-model activity on its platform surged during 2025, with Chinese models taking a 41% plurality of recent downloads. The larger point is less triumphal: parti…
Started by Hugging Face WatchMoonshotAI is transitioning Kimi CLI into Kimi Code CLI, its next-generation terminal agent. The important bit for existing users is practical: installing Kimi Code CLI automatica…
Started by Moonshot AI/Kimi WatchQwen Code is positioning itself as an open-source coding agent that can move from the terminal to IDEs, desktop apps, scripts and chat platforms. The interesting bit is not one fl…
Started by Alibaba Qwen WatchOpenAI intends to stop supplying its models to Cursor on 12 November after SpaceX acquired the coding company. Developers can keep using them during the notice period, but OpenAI …
Started by OpenAI WatchOpenAI says it will wind down the contract through which it provides models to Cursor after Cursor's acquisition by SpaceX. The immediate decision belongs to two companies, b…
Started by OpenAI WatchCoding agents can produce more code than ever, but Andrew Ng's latest letter argues that the person steering them still needs solid software engineering fundamentals. The typ…
Started by Andrew Ng WatchSpeech-recognition leaderboards are only useful when the languages on them resemble the languages people actually speak. Hugging Face has added Hindi, tagged hi-IN, to its Open AS…
Started by Hugging Face WatchLM Studio has shipped Bionic 1.1.0, and its most useful change is an attempt to make agent supervision less exhausting. Auto Review aims to stop safe, repetitive shell commands fr…
Started by LM Studio WatchvLLM 0.28.0 is a substantial serving-engine release rather than a version-number shuffle. The official notes count 584 commits from 270 contributors, including 76 first-time contr…
Started by vLLM WatchOpenAI has published its account of a security incident involving Hugging Face and set out what it says should happen next. The account matters less as a victory lap than as a...
Started by OpenAI WatchThe project has put a slower, stable release lane beside its breakneck build stream. Version 0.2.0, published on 21 August 2026, is the first release under a consistent semantic-v…
Started by llama.cpp WatchvLLM has made Model Runner V2 the default for all dense models and deleted the legacy PagedAttention implementation in v0.25.0, published on 11 July 2026. Pairing those changes in…
Started by vLLM Watch
