Select Language:
On April 21, a cutting-edge artificial intelligence startup in China announced the launch of its newest large language model, Kimi K2.6, part of the open-source Kimi series. This latest model aims to improve performance in programming, managing long-term tasks, and enabling multiple autonomous agents to work together effectively.
Kimi K2.6 is capable of maintaining continuous coding sessions for up to 13 hours. Its system supports a cluster of 300 sub-agents running simultaneously over 4,000 collaborative steps. These autonomous agents can execute persistent automated tasks and are compatible with popular frameworks such as OpenClaw and Hermes Agent.
These enhancements address previous limitations seen in earlier large models, which often generated only short snippets of code and struggled with complex software development. Earlier models often faced issues like losing context or producing inconsistent logic.
The upgrades to the Kimi series are part of a larger industry trend among leading AI companies worldwide. The focus is increasingly shifting from simply improving general capabilities to excelling in real-world task execution.
For example, other advanced AI systems can complete complex coding tasks quickly—such as finishing a 12.5-million-line codebase in just seven hours. Multi-agent collaboration architectures are also gaining traction, enabling AI to tackle more sophisticated research projects. Additionally, some models are now focused on handling long-range, multi-step tasks.
Alongside its technological progress, the company is actively pursuing new funding opportunities in both primary and secondary markets. Plans include an initial public offering in Hong Kong and a new funding round aiming to raise about one billion US dollars. Sources suggest that discussions around these fundraising efforts are ongoing, though the company has not publicly commented.
Following its latest funding round earlier this year, the company’s valuation reached approximately 10 billion dollars. However, its valuation still trails behind a listed competitor, which has a market capitalization of about 55.5 billion dollars as of the most recent market close. Reports indicate that the company may plan a pre-IPO fundraising round valued at around 18 billion dollars to strengthen its financial position.
ChatGPT
Perplexity
Gemini AI
Grok AI
