docs: refresh README recent progress (#125)

README.en.md

@@ -143,7 +143,7 @@ Choose one of the two quickstart paths first:
 
 If you want to try the OpenTalking + OmniRT + QuickTalk real-time digital-human path before setting up everything manually, use the community image we published on Compshare:
 
-- Image URL: <https://www.compshare.cn/images/TdDwmKZUZebI>
+- Image URL: [image link](https://www.compshare.cn/images/TdDwmKZUZebI?referral_code=Hid5KUhcqlZEptmMEwKy2F)
 - Exposed port: `5173` (WebUI; API traffic is proxied internally)
 - Guide: [Compshare image quick experience](docs/en/quick-start/compshare-image.md)
 
@@ -263,6 +263,18 @@ OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM
 
 ### Recent Progress
 
+- **2026-06-25: WeChat memory import and persona workflow**
+  Added WeChat memory persona import, documentation, and the related persona workflow. The frontend no longer treats persona selection and driving-model selection as mutually exclusive, so users can combine imported memory/persona context with the selected avatar driver.
+
+- **2026-06-23: Local CosyVoice TRT sidecar deployment**
+  Added the local CosyVoice sidecar deployment path with TensorRT / FP16 acceleration notes, runtime tuning, dedicated environment isolation, startup checks, and measured deployment guidance for pairing local TTS with QuickTalk.
+
+- **2026-06-22: Runtime configuration, memory refresh, and immersive scenes**
+  Added the runtime API configuration page, improved mem0 provider release during runtime refresh, and expanded the scene asset pipeline: scene asset APIs, asset-library integration, immersive conversation mode, scene/avatar anchoring, transparent background handling, and realtime media preservation across view switches.
+
+- **2026-06-18/19: Quickstart split, LightRAG runtime config, and scenario guides**
+  Split the quickstart into Compshare image and self-deployment paths, added LightRAG runtime configuration and quickstart updates, fixed dependency notes for mem0 / Hugging Face download tooling, and added the Huangshan digital-human guide.
+
 - **2026-06-12: QuickTalk local asset fixes and Apple Silicon support**
   Organized QuickTalk local weights, HuBERT, InsightFace paths, missing-asset checks, cache preparation, and health checks. Added Apple Silicon deployment docs for validating `quicktalk-cpu` with MPS / CPU on macOS arm64.
 

README.md

@@ -143,7 +143,7 @@ Choose one of the two quickstart paths first:
 
 If you want to try the OpenTalking + OmniRT + QuickTalk real-time digital-human path before setting up everything manually, use the community image we published on Compshare:
 
-- Image URL: <https://www.compshare.cn/images/TdDwmKZUZebI>
+- Image URL: [image link](https://www.compshare.cn/images/TdDwmKZUZebI?referral_code=Hid5KUhcqlZEptmMEwKy2F)
 - Exposed port: `5173` (WebUI; API traffic is proxied internally)
 - Guide: [Compshare image quick experience](docs/en/quick-start/compshare-image.md)
 
@@ -263,6 +263,18 @@ OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM
 
 ### Recent Progress
 
+- **2026-06-25: WeChat memory import and persona workflow**
+  Added WeChat memory persona import, documentation, and the related persona workflow. The frontend no longer treats persona selection and driving-model selection as mutually exclusive, so users can combine imported memory/persona context with the selected avatar driver.
+
+- **2026-06-23: Local CosyVoice TRT sidecar deployment**
+  Added the local CosyVoice sidecar deployment path with TensorRT / FP16 acceleration notes, runtime tuning, dedicated environment isolation, startup checks, and measured deployment guidance for pairing local TTS with QuickTalk.
+
+- **2026-06-22: Runtime configuration, memory refresh, and immersive scenes**
+  Added the runtime API configuration page, improved mem0 provider release during runtime refresh, and expanded the scene asset pipeline: scene asset APIs, asset-library integration, immersive conversation mode, scene/avatar anchoring, transparent background handling, and realtime media preservation across view switches.
+
+- **2026-06-18/19: Quickstart split, LightRAG runtime config, and scenario guides**
+  Split the quickstart into Compshare image and self-deployment paths, added LightRAG runtime configuration and quickstart updates, fixed dependency notes for mem0 / Hugging Face download tooling, and added the Huangshan digital-human guide.
+
 - **2026-06-12: QuickTalk local asset fixes and Apple Silicon support**
   Organized QuickTalk local weights, HuBERT, InsightFace paths, missing-asset checks, cache preparation, and health checks. Added Apple Silicon deployment docs for validating `quicktalk-cpu` with MPS / CPU on macOS arm64.
 

README.zh.md

@@ -143,7 +143,7 @@ OpenTalking 的 **编排层**（API / Worker / 前端）和 **数字人合成后
 
 如果你只是想先体验 OpenTalking + OmniRT + QuickTalk 的实时数字人链路，可以直接使用我们在优云智算发布的社区镜像：
 
-- 镜像地址：<https://www.compshare.cn/images/TdDwmKZUZebI>
+- 镜像地址：[镜像链接](https://www.compshare.cn/images/TdDwmKZUZebI?referral_code=Hid5KUhcqlZEptmMEwKy2F)
 - 对外端口：`5173`（WebUI，内部自动代理 API）
 - 操作文档：[优云智算镜像快速体验](docs/zh/quick-start/compshare-image.md)
 
@@ -263,6 +263,15 @@ OpenTalking 的 LLM 走 OpenAI-compatible 接口，把 `OPENTALKING_LLM_BASE_URL
 
 ### 近期进展
 
+- **2026-06-25：微信记忆导入与角色工作流**
+  新增微信记忆角色导入能力、配套操作文档和角色工作流；前端不再把 persona 选择和驱动模型选择视为互斥，用户可以把导入的记忆 / 角色上下文与当前选择的 Avatar 驱动模型组合使用。
+
+- **2026-06-23：本地 CosyVoice TRT sidecar 部署**
+  补齐本地 CosyVoice sidecar 部署路线，包含 TensorRT / FP16 加速说明、运行时参数调优、独立环境隔离、启动检查，以及本地 TTS 搭配 QuickTalk 的实测部署说明。
+
+- **2026-06-22：运行时配置、记忆刷新与沉浸式场景**
+  新增 runtime API 配置页，优化 runtime refresh 时 mem0 provider 的释放逻辑，并扩展场景资产链路：场景资产 API、资产库集成、沉浸式对话模式、场景 / Avatar 锚点、透明背景处理，以及切换视图时的实时媒体保持。
+
 - **2026-06-12：QuickTalk 本地资产修复与 Apple Silicon 支持**
   梳理 QuickTalk 本地权重、HuBERT、InsightFace 等资产路径，补充缺失资产检查、缓存准备和健康检查；新增 Apple Silicon 部署文档，支持在 macOS arm64 上使用 `quicktalk-cpu` 与 MPS / CPU 路线验证。