Moonshot AI’s July 2026 launch of the 2.8-trillion-parameter Kimi K3 MoE model—with its 1M-token context, native vision, and open weights—has set expectations for a rapid successor, yet no K4 announcement or release has occurred as of mid-September. Traders cite the company’s July push for additional Nvidia Blackwell chips to train a larger K4, alongside U.S. export-control scrutiny and allegations of model distillation, as key constraints slowing progress. Recent September updates routing some K3 traffic to a lighter K2.8 Preview for efficiency further signal internal focus on optimizing the current flagship rather than scaling to the next generation. Competitive pressure from Claude Fable 5 and GPT-5.6 Sol, plus Moonshot’s history of frequent iterations, keeps near-term odds low while December timelines remain more plausible if compute access improves.
基於Polymarket數據的AI實驗性摘要。這不是交易建議,也不影響該市場的結算方式。 · 更新於$17,851 交易量
2026 年 10 月 31 日
4%
2026年12月31日
6%
$17,851 交易量
2026 年 10 月 31 日
4%
2026年12月31日
6%
Kimi K4 refers to a product explicitly named Kimi K4 (e.g., Kimi K4.0 would count), or one that is recognized as the new flagship model or a successor to Kimi K3.0, consistent with the progression from Kimi K2 to Kimi K3. The release of any model within the Kimi K4 family will qualify as “Kimi K4”.
Products labeled as Kimi K3.5 or similar incremental versions will not count for this market’s resolution.
A qualifying model must be launched and publicly accessible, including via open beta or open rolling waitlist signups. A closed beta or any form of private access will not suffice.
The release must be either clearly defined and publicly announced by Moonshot AI as being accessible to the general public or otherwise made publicly accessible and explicitly labeled within the company’s official website. Labeling errors, placeholder text, or version names displayed on the website that do not correspond to a model that is actually accessible to the general public will not qualify.
The primary resolution source for this market will be official information from Moonshot AI, with additional verification from a consensus of credible reporting.
市場開放時間: Jul 24, 2026, 6:02 PM ET
Kimi K4 refers to a product explicitly named Kimi K4 (e.g., Kimi K4.0 would count), or one that is recognized as the new flagship model or a successor to Kimi K3.0, consistent with the progression from Kimi K2 to Kimi K3. The release of any model within the Kimi K4 family will qualify as “Kimi K4”.
Products labeled as Kimi K3.5 or similar incremental versions will not count for this market’s resolution.
A qualifying model must be launched and publicly accessible, including via open beta or open rolling waitlist signups. A closed beta or any form of private access will not suffice.
The release must be either clearly defined and publicly announced by Moonshot AI as being accessible to the general public or otherwise made publicly accessible and explicitly labeled within the company’s official website. Labeling errors, placeholder text, or version names displayed on the website that do not correspond to a model that is actually accessible to the general public will not qualify.
The primary resolution source for this market will be official information from Moonshot AI, with additional verification from a consensus of credible reporting.
Moonshot AI’s July 2026 launch of the 2.8-trillion-parameter Kimi K3 MoE model—with its 1M-token context, native vision, and open weights—has set expectations for a rapid successor, yet no K4 announcement or release has occurred as of mid-September. Traders cite the company’s July push for additional Nvidia Blackwell chips to train a larger K4, alongside U.S. export-control scrutiny and allegations of model distillation, as key constraints slowing progress. Recent September updates routing some K3 traffic to a lighter K2.8 Preview for efficiency further signal internal focus on optimizing the current flagship rather than scaling to the next generation. Competitive pressure from Claude Fable 5 and GPT-5.6 Sol, plus Moonshot’s history of frequent iterations, keeps near-term odds low while December timelines remain more plausible if compute access improves.
基於Polymarket數據的AI實驗性摘要。這不是交易建議,也不影響該市場的結算方式。 · 更新於



警惕外部連結哦。
警惕外部連結哦。
Frequently Asked Questions