Notice on the Offline of DeepSeek-V4-Flash-0731 Model

With the upgrade and iteration of DeepSeek's large language models, CloudBase will phase out and upgrade older models on the platform.

According to the adjustment plan, the DeepSeek-V4-Flash-0731 model (model parameter: deepseek-v4-flash-0731) on CloudBase AI will be officially taken offline starting from November 2, 2026, 00:00 (Beijing Time). Access to this model will no longer be available after that time.

Alternative Solution

We recommend migrating to the more powerful DeepSeek-V4.1-Flash model (model parameter: deepseek-v4.1-flash).

Migration Guidance

Please check and update the model name (model parameter) in your code before November 2, 2026 to complete the model switch and avoid any impact on your business operations.

Special Note: If you have not completed the migration by then, the system will automatically switch to the DeepSeek-V4.1-Flash model to ensure uninterrupted service. After the automatic switch, related calls will be billed according to the DeepSeek-V4.1-Flash model pricing.

Additional Note: Users who were previously switched from the DeepSeek-V4-Flash preview model to the DeepSeek-V4-Flash-0731 model will be automatically switched to the DeepSeek-V4.1-Flash model again when this model expires. No additional action is required.

---

Thank you for your understanding and support. CloudBase will continue to iterate its model product matrix to provide you with more powerful model capabilities and services.

If you encounter any issues, please visit the CloudBase Community for assistance.