[v0.6.1] N.E.K.O. (Macbook Support)

中文Change log

[主要更新]

• 支持多屏幕。
• 新增Macbook支持。
• 新增翻译字幕。当对话语言与系统语言不同时，弹出翻译提示。
• 移除底层傲娇属性。

[次要更新]

• 更好的初始化逻辑，降低首轮语音加载时间。
• 更好的热重置逻辑，降低热重置期间不回应语音的概率。
• 更好的TTS打断逻辑。不会再遗漏句子。
• 自定义语音（Cosyvoice）现在会自动预热。

[Bug修复]

• 修复了Live2D参数编辑器返回问题。
• 修复了内部更新失败提示。
• 修复了某些情况下语音模式开启主动视觉无法触发主动截图的问题。

English Change Log

[Major Updates]

• Added support for multiple displays.
• Added MacBook support.
• Added translated subtitles. When the conversation language differs from the system language, a translation prompt will pop up.
• Removed the systemetic Tsundere personality.

[Minor Updates]

• Improved initialization logic to reduce first-round voice loading time.
• Improved hot reset logic to reduce the chance of no voice response during hot resets.
• Improved TTS interruption handling; sentences will no longer be skipped.
• Custom voices (CosyVoice) are now automatically pre-warmed.

[Bug Fixes]

• Fixed the return issue in the Live2D parameter editor.
• Fixed incorrect internal update failure notifications.
• Fixed an issue where proactive screenshots could not be triggered in certain cases when voice mode and proactive vision were enabled.

New Contributors

• @coderabbitai[bot] made their first contribution in #199
• @wsqstar made their first contribution in #195
• @TL0SR2 made their first contribution in #211

Full Changelog: v0.6.0...v0.6.1

[v0.6.0] N.E.K.O. (Proactive Era)

中文Change log

[主要更新]

• 新增主动视觉功能（默认关闭）。猫娘会在主动搭话时获取屏幕信息，或在语音对话中时不时窥屏。
• 更加完善的自定义API支持。现在你可以使用自定义API来自由搭配服务商。你可以使用免费API搭配阿里的TTS语音。
• 新增基于循环神经网络的主动降噪系统，有助于提升语音识别表现。

[次要更新]

• 更新了阿里对话模型Qwen3-Omni-1201的默认参数，缓解复读问题。
• 文本模式下新增了一系列复读检测，触发时会自动清除复读对话并提醒用户对策。
• 新增插件系统（处于实验状态）。
• 语音克隆时可以选择参考语音的语言了。
• 使用Ogg Opus而非PCM传输TTS数据，显著降低带宽压力。
• 使免费/Step模型的文本模式下同样支持Web Search。
• Docker容器新增nginx反向代理。

[Bug修复]

• 修复了UI界面中一系列点击事件和状态重置问题。
• 当猫娘被打断时，不会再出现重复的记忆块。
• 麦克风静音检测不再计算语音系统准备期间的时间。

English Change Log

[Major Updates]

• Added Active Vision feature (disabled by default). The catgirl can capture screen information when initiating conversations, or occasionally peek at the screen during voice chats.
• Improved Custom API support. You can now freely mix and match service providers using custom APIs. For example, you can pair a free API with Alibaba’s TTS service.
• Introduced an active noise reduction system based on recurrent neural networks, helping to improve speech recognition performance.

[Minor Updates]

• Updated the default parameters for Alibaba’s dialog model Qwen3-Omni-1201, reducing repetition issues.
• Added a series of repetition-detection mechanisms in text mode. When triggered, repeated dialogue will be automatically cleared and users will be notified with suggested countermeasures.
• Added a plugin system (experimental).
• Voice cloning now allows selecting the language of the reference audio.
• Switched TTS data transmission from PCM to Ogg Opus, significantly reducing bandwidth usage.
• Enabled Web Search in text mode for free/Step models as well.
• Added an nginx reverse proxy to the Docker container.

[Bug Fixes]

• Fixed a number of click-event and state reset issues in the UI.
• Prevented duplicate memory blocks from appearing when the catgirl is interrupted.
• Microphone mute detection no longer counts the system’s voice preparation time.

What's Changed

• Docker封装，部署和启动流程的优化。Docker部署文档更新 by @WUHINS in #157
• fix: prevent click-through in advanced settings and improve UX by @Tonnodoubt in #158
• 修复免费版本 UI 显示和保存后状态显示错误 by @Chiya-mt in #159
• Added noise reduce funcionality with RNNoise. by @wehos in #168
• 添加可以根据屏幕内容主动搭话的功能 by @Chiya-mt in #160
• fix: 在加载新模型和刷新页面时关闭设置窗口 by @CN-QanYi in #165
• Add user-facing plugin subsystem (plugin server, registry, runtime state) by @wislap in #161
• Added repetition detection and notification. by @wehos in #170
• Support custom API. Screenshot utils unified usage. by @wehos in #171
• Refactor main_server. Adjust build profiles. by @wehos in #173
• 解决了自定义API设置视觉模型后模型选择错误的问题 by @Chiya-mt in #174
• feat(交互): 在拖拽模型时临时禁用按钮事件拦截 by @CN-Zephyr in #175
• feat(主动对话):基于标题搜索的主动对话功能 & 将热门内容改为首页推荐并更新相关逻辑 by @CN-QanYi in #176
• 新增语音聊天时如果开启了主动视觉则每十秒共享一次屏幕（1帧画面） by @Chiya-mt in #179
• feat(系统路由): 添加AI回复内容验证规则 by @CN-Zephyr in #181
• Enhance denoise and AGC quality for better VAD. by @wehos in #184
• 修复语音或屏幕共享按钮状态异常的问题 by @Chiya-mt in #189

Full Changelog: 0.5.5...v0.6.0

[v0.5.5] N.E.K.O. (Agent-enhanced)

中文Change log

[主要更新]

• 对Agent前端进行了大量翻新。现在Agent功能（键鼠&MCP）重新恢复支持。
• 新增了Live2D参数调整（捏人系统），可以对模型的发饰、服饰、颜色等数据进行直接编辑
• 锁定状态改为游戏模式，不再弹出菜单栏

[次要更新]

• 新增了可以改色的兽耳娘模型(credit: umijin68)
• 阿里的对话模型已更新至Qwen3-Omni-1201最新版本
• 现在任何状态下拖拽Live2D模型都会立即保存位置
• 美化了前端图标，使用更美观的Icon以及呼吸灯特效
• 现在菜单栏大小会自适应Live2D模型大小
• 人设编辑变为多行输入模式，方便撰写小作文

[Bug修复]

• 修复了文本模式下，热重置意外使用语音session的bug
• 修复了tts session在结束后没有手动结束的bug
• 修复了一些与鼠标穿透有关的bug
• 将端口监听由0.0.0.0改为127.0.0.1回路，提高了安全性；手机直连用户请注意这一点

English Change Log

[Major Updates]

• Performed extensive revamps to the Agent frontend. Agent features (keyboard/mouse & MCP) are now supported again.
• Added Live2D parameter customization (character-tuning system), allowing direct editing of model accessories, outfits, colors, and other data.
• In locked mode, the interface now switches to Game Mode, and the menu bar will no longer pop up.

[Minor Updates]

• Added a recolorable beast-ear girl Live2D model (credit: umijin68).
• Updated Alibaba's dialogue model to the latest Qwen3-Omni-1201 version.
• Dragging the Live2D model will now immediately save its position in any state.
• Improved frontend icons with more aesthetically pleasing designs and a breathing-light effect.
• The menu bar size now auto-adjusts based on the Live2D model size.
• Character profile editing now supports multi-line input, making it easier to write longer descriptions.

[Bug Fixes]

• Fixed a bug where hot-reload accidentally used the voice session while in text mode.
• Fixed an issue where the TTS session did not properly terminate after ending.
• Fixed several mouse-through (click-through) related bugs.
• Changed port listening from 0.0.0.0 to 127.0.0.1 loopback to improve security — mobile device direct-connection users, please take note of this.

What's Changed

• Add i18n support for parameter editor and improve persistent expression handling by @Tonnodoubt in #132
• Liangfix: Fix Live2D model disappearing immediately after loading by @Tonnodoubt in #136
• 为主人档案和猫娘档案的属性添加滚动条，方便阅读和修改 by @Chiya-mt in #137
• Critical update to enhance security and robust session handling. by @wehos in #138
• feat(交互): 添加模型位置和缩放的自动保存功能 by @CN-QanYi in #139
• Fix: Add error handling for motionManager.update and coreModel.update overrides by @Tonnodoubt in #140
• 修复自定义API模式下模型ID不保存与不显示填写内容的问题；修复模型设置页UI问题 by @Chiya-mt in #142
• fix: 修复Live2D参数重置逻辑错误 by @CN-QanYi in #143
• 修复了模型设置页中切换角色不点保存也会保存的问题，并添加了未保存提示 by @Chiya-mt in #146
• 参数自定义界面点左上角退出的时候会重新打开一个主页，而不是回到原来的主页 by @CN-Zephyr in #150
• feat: 添加用户偏好设置加载功能以增强模型个性化 by @CN-QanYi in #151
• feat(agent): 实现弹窗状态机并优化Agent控制逻辑 by @CN-QanYi in #153
• Live2d refactor into several modules by @wehos in #152
• Added more i18n config. Fixed parameter-motion interaction logic. Improve agent menu synchronization logic. by @wehos in #155

Full Changelog: v0.5.4...0.5.5

[v0.5.4] N.E.K.O. Steam版

What's Changed

• 集成Steamworks API。支持创意工坊和成就系统。
• 重写的Agent模式，更快的响应速度。
• 更加精简的依赖。现在发行版大小仅 250MB 。
• 跨服务器同步由多进程改为多线程，提高了启动效率。
• 调整了前端加载时序，有效减少了前端竞态问题。
• 修复了一系列i18n bug，更新了API页面的说明。
• 修改了初始人设、温度、系统设定，显著提升了对话质量。
• 为所有LLM推理添加了有条件的重试机制。
• 构建了Docker容器，可以在远程服务器/NAS上部署。
• 改善了UI主题和样式，请她离开后可以拖拽。

此外修复了一百个Bug，已经写不动日志了。以下内容为自动生成，仅包含了PR部分。

• Steamworks sdk added by @wehos in #89
• Split live2d script. Improve live2d pointer event logic. by @wehos in #90
• fix(交互): 修复按钮拖动与点击的冲突问题并优化返回按钮定位 by @CN-QanYi in #88
• feat: Add global window management to prevent duplicate window opening by @Tonnodoubt in #92
• Upgrade LangChain to 1.0: Simplify dependencies and improve performance by @Tonnodoubt in #68
• Update steam branch to latest main. by @wehos in #93
• 当修改猫娘名称时猫娘记忆名称也一同修改 by @Chiya-mt in #94
• fix: Live2D 模型管理器进行了改进，重点在于用户偏好设置的持久化以及更佳的默认模型屏幕位置 by @CN-QanYi in #95
• refactor(Live2DManager): 移除自动保存功能并修复复位位置逻辑 by @CN-QanYi in #98
• 修复Agent开关项多次触发bug by @Chiya-mt in #97
• Auto-initialize memory server for new characters and enable real-time chat recording by @Tonnodoubt in #99
• 修复构建时的python依赖文件的复制路径 by @WUHINS in #100
• 优化Live2D情绪系统和日志输出 by @Chiya-mt in #102
• Fixed Workshop page style, ui logic, remove hard coded dir etc. by @nantxcel in #101
• 确保i18n文件完全遵循了正确的模板变量格式，使用“{{}}”而不是“{}” by @Chiya-mt in #103
• feat(ui): 为任务HUD添加拖拽和折叠功能 by @CN-Zephyr in #106
• Update entrypoint.sh，以健全docker容器启动，docker容器内置启动脚本 by @WUHINS in #109
• Update Dockerfile: Migrate to single-stage Debian-based build by @WUHINS in #111
• 用预构建镜像替换neko-main构建并简化docker编排 by @WUHINS in #112
• fix: Prevent text chat interruption when switching other characters' voice by @Tonnodoubt in #110
• Merge steam i18n to main by @wehos in #114
• Updated i18n and minor fixes for steam. by @wehos in #117
• docker编排镜像拉取标签更新 by @WUHINS in #118
• 上传模型时将动作文件中关于口型的部分都设置为空值 by @Chiya-mt in #122
• feat(LLMSessionManager): 支持角色切换时的音色热更新 by @CN-QanYi in #123
• fix(agent): 修复频繁切换Agent开关导致的竞态问题 by @CN-QanYi in #124
• feat(会话管理): 添加系统准备状态通知机制 by @CN-QanYi in #127
• Update steam achievements and workshop EULA by @wehos in #128

New Contributors

• @WUHINS made their first contribution in #100

Full Changelog: v0.5.3...v0.5.4

[v0.5.3] N.E.K.O.

What's Changed

提升了性能，添加了Win证书（不会再有未知发行者提示），增加了多语言适配（i18n）和开机自启动功能。修复了一百个Bug，已经改不动日志了。接下来进入Steam发售前的封闭开发期，不会再有Release。

• 调整API Key设置页面，可加入自定义API by @CN-QanYi in #40
• Enhance Agent menu with real-time availability checking and improved UI by @Tonnodoubt in #41
• 解决了克隆并设置声音后无法切换为不使用克隆声音的问题 by @Chiya-mt in #42
• 猫娘档案的展开状态记录到localstorage以便刷新后自动展开 by @noahzaozao in #43
• Auto-delete memory files when deleting catgirl character by @Tonnodoubt in #44
• [FIX] 确保 lanlan_config.lanlan_name 更新到 chara_manager.html 当前选中的猫娘 by @noahzaozao in #47
• feat(交互): 新增"请她回来"功能并优化界面状态管理 by @CN-QanYi in #51
• 为角色设置页添加保存与取消按钮的显示逻辑使有修改后才显示 by @Chiya-mt in #52
• feat(ui): 统一按钮图标样式 by @CN-Zephyr in #55
• Fix mobile menu layout by @noahzaozao in #56
• Fix mobile chatbox layout by @noahzaozao in #57
• Migrate API providers and default models from hardcoded to JSON config by @Tonnodoubt in #58
• 将聊天框最小化背景回调为透明 by @Chiya-mt in #59
• 为live2d模型设置页添加全屏按钮 by @Chiya-mt in #64
• Complete i18next internationalization and optimize code structure by @Tonnodoubt in #62
• 回调部分对index.html的更改修复bug by @Chiya-mt in #66
• Fix UI state synchronization issues when leaving and returning session by @Tonnodoubt in #67
• style: update input style in voice_clone.html by @xxynet in #73
• Fixed i18n functionality and optimized notification messages by @noahzaozao in #72
• UI Improve chat dialog button styling with custom icons and color scheme by @Tonnodoubt in #76
• feat(API配置): 为自定义API添加模型ID配置项 by @CN-QanYi in #77
• Fix i18n params issue with incorrect {} usage by @noahzaozao in #78
• fix: sync focusModeEnabled when returning from l2d page by @noahzaozao in #80
• fix live2d import default position by @noahzaozao in #82
• fix(api-settings): Improve API Key settings page loading logic and error handling by @noahzaozao in #79
• Fix settings popup checkbox visual state sync issue by @noahzaozao in #85
• fix(voice-clone): Fix timeout retry logic and improve error handling by @Tonnodoubt in #84
• 解决API设置页配置可能读取错误的问题 by @Chiya-mt in #83
• fix: Improve microphone button state management and prevent duplicate… by @Tonnodoubt in #86
• 修复猫娘重影问题。修复窗口双重打开问题。
• 通过恢复GPU整体渲染提升了性能表现。

New Contributors

• @CN-Zephyr made their first contribution in #55
• @xxynet made their first contribution in #73

Full Changelog: v0.5.2...v0.5.3

[v0.5.2] 正式更名为N.E.K.O.，UI大升级！

我们对UI进行了一系列重大升级。Steam商店页已经在审核中！

What's Changed

• [Important] UI remade, with a lot of art resources and visual improvement. Rebranded to N.E.K.O.
• [UI Improvement] Autoselecting catgirl in memory review
• [UI Improvement] Autoselecting catgirl in memory review.
• [UI Improvement] Auto-filling core api key when setting up sup-api.
• [UI Improvement] Auto-hiding the current l2d model when setting up new models.
• [UI Improvement] Disable auto-folding when setting up catgirl property.
• [UI Improvement] Auto-reloading catgirl when l2d is changed.
• [Minor Improvement] More improvements and bug fixes with UI logics .

中文版

• [重要更新] UI再次重制，添加了大量美术资源并进行了视觉改进。名称重制为N.E.K.O.
• [界面优化] 在记忆回顾中自动选择猫娘。
• [界面优化] 在设置 sup-api 时自动填充核心 API 密钥。
• [界面优化] 在设置新模型时自动隐藏当前的 L2D 模型。
• [界面优化] 在设置猫娘属性时禁用自动折叠。
• [界面优化] 当 L2D 模型更改时自动重新加载猫娘。
• [优化] 进一步优化 UI 逻辑并修复了一些错误。

New Contributors

• @CN-QanYi made their first contribution in #32
• @nantxcel made their first contribution in #34

Full Changelog: v0.5.1...v0.5.2

[v0.5.1] Lanlan最终版

免费模型已上线！文本输入和主动对话已得到支持。这是项目更名为Project N.E.K.O.并转为LGPL协议前的最后一个版本。预计下周上架steam！

What's Changed

• [Important] UI completely remade.
• [Important] Launcher no longer needed for the one-click package.
• [Important] Proactive mode added. Catgirl can now proactively engage in conversation.
• [Functionality] Monitor server added for cross-device sync.
• [Functionality] Store voice IDs for reuse.
• [Functionality] Support vision for non-visual model (e.g., free, Step) and text input mode.
• [Functionality] Live2D can now be dragged to resize on mobile devices.
• [Bug fixed] Fixed one issue with hot swap
• [Minor Improvement] Added icon for the webpage.
• [Minor Improvement] Change system prompt to reduce repetition.
• [Minor Improvement] Autofolding dialogue window.

中文版

• [重要更新] UI完全重置，带来全新的用户体验！
• [重要更新] 一键包不再包含外置启动器。现在可以真·一键启动。
• [重要更新] 新增主动对话模式，允许猫娘主动搭话。
• [新功能] 观察者服务器现在已可用，提供跨设备同步输出的功能。
• [新功能] 声音ID会被储存以供复用。
• [新功能] 非视觉模型也可以支持视觉。文本输入模式新增截图功能。
• [新功能] 手机上Live2D模型现在可以被拖拽缩放。
• [Bug修复] 修复了热切换导致的一个Bug。
• [优化] 网页现在有Icon图标了。
• [优化] 通过调整系统提示词，减少了复读的情况。
• [优化] 文本输入框现在会自动折叠。

New Contributors

• @jihu33 made their first contribution in #22
• @Tonnodoubt made their first contribution in #26

Full Changelog: v0.5.0...v0.5.1

[v0.5.0] 多端全能AI伙伴（零配置一键运行）

免费模型已上线！文本输入和硅基流动已得到支持。下一步我们想要：
完善长期记忆，完善手机端的对话管理和双端同步，11月上架steam！

What's Changed

• [Important] Added a fully functional free model (exclusive for Lanlan use, credited to Step.ai).
• [Important] Added text input support. Support Deepseek.
• [Important] API Key, Memory, and Log files all synced to My Documents automatically.
• [Functionality] User audio transcription is now displayed at the frontend.
• [Bug fixed] Changed audio temp file server to a domestic server site.
• [Minor Improvement] Qwen3 no longer speaks '.' to indicate silence.
• [Minor Improvement] Better logs, frontend error presentation, and interruption logic.

中文版

• [重要更新] 提供功能齐全的免费模型（由阶跃星辰提供的，本项目特供福利）。
• [重要更新] 支持文本输入和硅基流动（Deepseek）。
• [重要更新] API设置、记忆和日志文件全部迁移至我的文档。这是为了未来上架steam。
• [新功能] UI调整，用户输入的语音转录也会在前端显示。
• [Bug修复] 修复了临时文件服务器被墙导致自定义语音无法上传的问题。
• [优化] 模型不再输出句号以示沉默。
• [优化] 更好的日志，前端错误显示，用户打断逻辑。

Full Changelog: v0.4.1...v0.5.0

[v0.4.1] 多端全能AI伙伴（3分钟上手）

What's Changed

• [Important] MCP Client enabled. Please refer to the MCP Router for setting up the server.
• [Important] GPT-Realtime, Step-Audio-2, and Qwen-3-Omni are supported. Default core model changed to qwen3-omni-realtime.
• [Functionality] Microphone selection supported.
• [Bug fixed] Fix a name conversion issue with the initial prompt that leads to a performance drop.

New Contributors

• @xiaoniao427 made their first contribution in #18

中文版

• [重要更新] 支持MCP客户端。请通过MCP Router设置MCP 服务器。
• [重要更新] 支持GPT-Realtime，Step-Audio-2和Qwen3-Omni。默认模型替换为qwen3-omni-realtime.
• [新功能] 支持在App（及网页）内选择麦克风。
• [Bug修复] 修复了一个造成性能下降的初始提示词名称转换问题。

Full Changelog: v0.4.0...v0.4.1

[v0.4.0] 多端全能AI伙伴（3分钟上手）

Lanlan can now use a mouse and keyboard! 兰兰可以使用键盘和鼠标了！.
Due to dependency updates, we do not provide a progressive update patch for this version.

What's Changed

• [Important] Agent module added. Computer use enabled.
• [Functionality] Support persistent expression to help remove live2d watermarks.
• [Functionality] Updated lipsync logic to cover more live2d models.
• [Functionality] New main entrance based on rust.

New Contributors

• @CarrotFree made their first contribution in #16

中文版

• [重要更新] 新增智能体服务器。兰兰现在可以接管键盘鼠标了。
• [新功能] 通过持久化表情，支持live2d去水印。
• [新功能] 更新唇形同步逻辑，支持更多live2d模型。
• [新功能] 新的由rust语言编写的主入口。

Full Changelog: v0.3.2...v0.4.0