Kimi raises $2B, valuation quadruples to $20B in six months
In court, OpenAI president Greg Brockman revealed that Elon Musk demanded 51% control of the startup to fund a Mars city — and stormed out when they refused.
In court, OpenAI president Greg Brockman revealed that Elon Musk demanded 51% control of the startup to fund a Mars city — and stormed out when they refused.
Mistral AI released Mistral Medium 3.5, a 128B parameter open model that outperforms Claude Sonnet 4.5 on agent benchmarks, along with cloud updates to Mistral Vibe and Le Chat.
Asking in Chinese costs Claude 65% more tokens than English. OpenAI only 15% extra. The culprit? Tokenization efficiency depends on language representation in training data.
Moonshot's Kimi K2.6 became the most-used model on OpenRouter within a week of launch, consuming 1.88 trillion tokens and surpassing Claude Sonnet 4.6 by nearly 40%.
Leaked specs suggest Kimi's upcoming K3 model will have over 2.5 trillion parameters and a 1M token context window, but the bottleneck isn't technology—it's compute.
The State Department has sent diplomatic cables to embassies worldwide, warning that Chinese AI companies DeepSeek, Moonshot AI (Kimi), and MiniMax are using 'distillation' to steal US AI models. Beijing calls the allegations 'groundless'.
Moonshot AI has open-sourced a verification tool to ensure third-party APIs running its Kimi models match the official performance, addressing a major quality control issue in the open-source AI ecosystem.
Moonshot AI has open-sourced its Kimi K2.6 model, claiming its coding performance now rivals closed-source leaders like GPT-5.4 and Claude Opus 4.6.
Moonshot AI's new Kimi K2.6-code-preview is a trillion-parameter model built for programming, with benchmarks suggesting it performs on par with Anthropic's Claude Sonnet 4.6.