Kimi K2.8 Preview model fully launched on Kimi Code
月之暗面 Kimi K2.8 Preview 模型全量上线 Kimi Code，综合性能接近 K3
For users of the Kimi Code platform, the K2.8 Preview model offers improved performance and a larger context window, making it a better option than the previous K2.7 Code version.
The Moonshot Kimi K2.8 Preview model has been fully released on the Kimi Code platform.
The model ID remains kimi-for-coding, so no changes are needed for clients or third-party tools.Officially, K2.
8 Preview's overall performance is close to K3, with improved thinking efficiency.Coding and agent capabilities are enhanced, and thinking efficiency is significantly better than K2.
7 Code.All membership tiers now support up to a 1M context window.
Kimi K2.8 Preview supports thinking effort adjustment, with low/high/max tiers matching K3's settings.
When thinking is disabled, both K3 series and K2.8 Preview requests are routed to the K2.
8 Preview (no thinking) version.
月之暗面 Kimi K2.8 Preview 模型全量上线 Kimi Code，综合性能接近 K3
IT之家 9 月 11 日消息， 月之暗面 Kimi K2.8 Preview 模型今日全量上线 Kimi Code ，Model ID 不变，仍为 kimi-for-coding，客户端与第三方工具无需修改配置即可使用。 官方表示， K2.8 Preview 综合性能接近 K3 ，思考效率更高。编码与智能体能力全面提升，思考（thinking）效率较 K2.7 Code 显著改善， 全部会员档位均开放最高 1M 上下文窗口 。 IT之家获悉，Kimi K2.8 Preview 支持 thinking effort 档位调节，与 K3 的思考等级一致，支持 low / high / max 三档（默认 max）。关闭 thinking 后，K3 系列与 K2.8 Preview 的请求均路由到 K2.8 Preview（无思考）。 相关阅读： 《 月之暗面开源 Kimi K3 模型，2.8 万亿参数 》
This report is categorized under AITOP's "Models" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to Chinese AI companies expanding and localizing across Southeast Asia and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Kimi, Kimi Code, K2.8, coding assistant.
Source attribution: curated from IT之家.
