China AI Native Industry Insights – 20261116 – DeepSeek | Qoder | Tencent | more

Today’s digest highlights notable advancements from DeepSeek and Tencent. The common theme linking these updates is enhanced multimodal AI capabilities extending across voice, web, and application interfaces. DeepSeek has unveiled a comprehensive suite of APIs incorporating web search, file input, and voice capabilities, while Tencent introduces the Hy Realtime voice model with innovative tool-calling and singing functions. Additionally, Qoder IDE’s version 1.40 brings new features like a task metrics display and plan mode model switching, offering developers improved session management and interface interaction. Moonshot AI provides a practical tutorial on building 3D websites with Kimi K3.1 Pro and Kimi Code. Discover more in Today’s China AI Native Industry Insights.

1. DeepSeek releases comprehensive suite of application APIs including web search, file input, and voice capabilities

DeepSeek has opened its full suite of application APIs, integrating features from the DeepSeek App directly into existing APIs. The V4.1-Flash model now supports web search, deep thinking search, file input for PDF and Office documents, URL reading, and safety moderation through simple parameter toggles in Messages, Responses, and Chat Completions interfaces. Underlying capabilities including Search, Fetch, OCR, ASR, and TTS are also available as standalone API endpoints for flexible integration. DeepSeek has released an open-source frontend demo and pricing structure for these new API tools to help developers replicate DeepSeek App functionality in their own applications.
Read more: https://mp.weixin.qq.com/s?__biz=Mzk0OTYwNzc3NQ==&mid=2247485910&idx=1&sn=529f50ba3b37e39319dad1faf7a122cf&chksm=c214c90857fe9a4263f6fd592017d28ec10a473c8742337d1a06b5a661ee48359430b69cbd81&scene=0&xtrack=1#rd


Video Credit: NotebookLM

2. Qoder IDE releases version 1.40 with task metrics display, IDE helper command, session tab status, and plan mode model switching

Qoder IDE announced the release of version 1.40 with several new features for AI-assisted development workflows. The update includes post-task metrics showing execution time and credit usage, a new /qoder-ide-helper command for adjusting settings and consulting features through conversation, session tab status indicators with keyboard shortcuts for switching between multiple concurrent agent tasks, and the ability to switch models between planning and execution phases in Plan mode. These updates target developers working with AI agents for coding tasks, aiming to improve workflow visibility and control.
Read more: https://mp.weixin.qq.com/s?__biz=MzE5ODI2NzI5Nw==&mid=2247494564&idx=1&sn=527124f4075f72ea23bcfb5269335f81&chksm=97f913c6896d64b3ea605d39285edc17d0df598c93c4f9805aa53ab32b718c68c1c1cfc7732b&scene=0&xtrack=1#rd


Video Credit: The original article

3. Tencent releases Hy Realtime voice model with tool-calling and singing capabilities

Tencent’s Hunyuan team announced the release of Hy Realtime, a new real-time voice model featuring advanced speech understanding, instruction following, and humanized expression capabilities. The model introduces Agentic tool-calling functionality and singing abilities developed in collaboration with Tencent Music’s TME Lingyin AI team. Hy Realtime is already deployed in products including Yuanbao voice calls and Tencent Customer Service, and will soon be available via Tencent Cloud API. The model demonstrates improvements in cross-modal alignment, low-latency voice generation, and multi-turn contextual understanding through enhanced pretraining and post-training frameworks.
Read more: https://mp.weixin.qq.com/s?__biz=MzkwODU2OTQyNQ==&mid=2247498965&idx=1&sn=fc0a8fa819b333cf16814ae46d6ccd4f&chksm=c1bede2b6a953e3a146eb7ff1b0c10f81c56d0ada2eb8ecd46c7f06c36e6c791c0b92ca33964&scene=0&xtrack=1#rd


Video Credit: The original article

4. Moonshot AI releases tutorial on creating 3D websites using Kimi K3.1 Pro and Kimi Code

Moonshot AI published a detailed tutorial demonstrating how to use Kimi K3.1 Pro and Kimi Code to create interactive 3D websites. The tutorial uses a ceramic studio landing page as a case study, showcasing how users can generate 3D models, animations, and interactive web experiences through natural language prompts. Kimi Code, available as a desktop application, IDE plugin, or CLI tool, allows users to create complex 3D web content by constructing code, calling professional tools, and integrating third-party resources. The tutorial highlights Kimi’s enhanced 3D capabilities and web design optimization introduced with the K3.1 Pro release.
Read more: https://mp.weixin.qq.com/s?__biz=MzkzMTY4NTIyNA==&mid=2247484433&idx=1&sn=088871c88998f37f1d5675cbc9822d68&chksm=c31590ef0cb6ddf70247bd70dc469e52ccaa613aa7d83078ef26f7633b23255ed9557680ac9b&scene=0&xtrack=1#rd


Video Credit: The original article

That’s all for today’s China AI Native Industry Insights. Join us at AI Native Foundation Membership Dashboard for the latest insights on AI Native, or follow our linkedin account at AI Native Foundation and our twitter account at AINativeF.

Blank Form (#4)
AI Native Foundation logo
[email protected]

About

Copyright 2026 AI Native Foundation© . All rights reserved.​