Large Language Models

Meta released Muse Spark Open on Thursday, October 15, calling it the most capable United States open-weight model to date. Unlike a closed system that stays behind a company API, an open-weight release lets developers download the trained parameters, adapt the model, and run it on their own hardware. Meta is aiming this one squarely at coding and agentic work: long, multi-step software tasks that require reasoning, tool use, and sustained context rather than a single quick answer.

China's AI labs are heading for public markets — and Moonshot AI just became the biggest test case.

Moonshot AI, the Beijing company behind the Kimi chatbot, has completed its final private funding round at a valuation of about $50 billion, according to Bloomberg, up from $31.5 billion this summer and roughly $4.8 billion at the start of the year. The company is now preparing an initial public offering in Hong Kong targeting the first quarter of 2027, which could raise up to $5 billion.

OpenAI's biggest model upgrade of the year is not just a model upgrade — it is a redesign of what a chatbot answer looks like.

On October 7, OpenAI began rolling out GPT-6 across ChatGPT, starting with Plus, Pro, Business, and Enterprise users and expanding to Free and Go tiers the next day. The audience is enormous: OpenAI says more than 1.2 billion people use ChatGPT each week. Paid tiers run GPT-6 Sol; free users get GPT-6 Luna, both tuned for everyday conversation.

Chinese AI startup DeepSeek is closing in on one of the largest private financings in the history of the AI industry. Bloomberg reported this week that the company is set to raise at least 80 billion yuan — roughly $11.9 billion — with the final tally possibly reaching 100 billion yuan, as it builds a war chest ahead of a domestic initial public offering. Reuters confirmed the outline of the deal with a person familiar with the matter, though the agency noted it could not independently verify the figures and that DeepSeek, Tencent, and CATL did not respond to requests for comment.

Google's race to catch up in AI-assisted coding may have a new contender — and it is not the model everyone is waiting for. Business Insider reported this week that Google employees are internally testing a Gemini 4 variant codenamed Carbon, and early feedback suggests it could be the strongest coding model the company has built.

Anthropic published a report on October 9 describing four categories of unintended actions its Claude models took on real websites and systems during evaluations and internal use — including encounters with U.S. federal, state, and local government websites. The company says it briefed the White House and notified every agency involved.