# cherry-studio: what changed from 1 to 2 - Product: cherry-studio (https://whatsnew.fyi/product/cherry-studio) - Vendor: cherry-studio - Range: changelog entries numbered after v1.9.13 up to and including v2.0.12, stable releases only - Entries below: 12 releases (newest first) - Resolved: 1 is v1.9.13 and 2 is v2.0.12, the newest stable release of each major we track - Carrying security changes: 2 · CVEs mentioned: 0 · Mentioning breaking changes: 0 · Removing or deprecating something: 1 - Page: https://whatsnew.fyi/product/cherry-studio/compare/1...2 What's New is an index, not a publisher: every entry below links to the vendor's own release notes, which are the authoritative source. Entries are labelled where they are hand-curated sample data, pre-releases, or drawn from a secondary source such as a developer blog. Reuse: the summaries, labels and curation here are © What's New. Quote freely with attribution and a link back; wholesale republication of the corpus is not permitted — terms: https://whatsnew.fyi/terms. The vendors' own release notes remain their publishers'. ## What changed (269 changes, grouped by kind) ### Added #### v2.0.12 (2026-09-04) - Settings gains a sidebar search to find any setting row across all sections including title, description, aliases, and Chinese pinyin, with scroll, highlight, and full keyboard navigation - Settings search now covers all sections with static settings rows, including MCP tabs - Translate model now has its own temperature, top-p, and reasoning effort settings, editable from the translation settings panel and translate page toolbar - In agent session messages and rendered artifact Markdown previews, workspace links open files in the artifact pane and directories in the system file manager, with relative links resolving from the agent workspace root #### v2.0.11 (2026-09-03) - Sign in to Cherry Cloud for entitlement-aware Agent models, automatic local Gateway startup, quota-aware model selection, and durable Product Session handling - Configure the compression trigger threshold per assistant or globally in Settings - Improved installation and navigation for Mini Apps with drag-and-drop packages, stable tabs and icons, and transparent Translate and Painting views - Press Ctrl+A (Cmd+A on macOS) in the file list to select all files in the current view - Added an inline delete button to remove a message's translation in one click - Hide the chat assistant from the sidebar via Appearance Settings, and launch directly to the first visible sidebar app - Added a global search action to the Agent conversation toolbar - Add Hy4 Preview, Ling 3.0 Flash, and Qwen3.8 Flash model support - Anthropic-compatible providers can now keep prompt caches for 1 hour from API Settings - Added initial Turkish localization support across Cherry Studio - Shows the available disk space when a backup or restore fails because the disk is full #### v2.0.10 (2026-08-28) - Show run status on task cards for scheduled tasks #### v2.0.9 (2026-08-24) - Enable remote catalog hot-update without an app release in provider registry - Publish August 2026 privacy policy update - Restore Ctrl+L clear messages shortcut - Add model settings navigation in assistant settings - Pin assistants and agents to the sidebar #### v2.0.8 (2026-08-19) - add skill global enablement controls - add floating task progress capsule to agents - add interactive screen capture with annotation and OCR - add TypeScript-guided code mode tools to pi-runtime - add layout-preserving PDF translation via BabelDOC - record translations in history and file manager for pdf-translation - surface MCP prompts and resources in the composer and to the model - add durable cross-session delivery to agent-session - allow private-network fetch targets behind a preference in remote-fetch #### v2.0.7 (2026-08-17) - Add sidebar help and diagnostic upload to help-feedback - Add auto permission mode for pi agents and flatten the runtime picker - Integrate DeepSeek Harness in code-mate #### v2.0.5 (2026-08-13) - Add built-in support agent - Add app update notification preference - Install a skill from a GitHub SKILL.md URL #### v2.0.4 (2026-08-13) - Add back-to-top button and line-break mode to notes - Add delete impact preview to agent workspace - Add tab navigation shortcuts - Restore MCP-over-HTTP endpoints (/v1/mcps) in API gateway - Add opt-in MCP sessions with server-to-client push and progress in API gateway - Handle GROUP_MESSAGE_CREATE event with per-channel mention-only toggle for QQ channel #### v2.0.3 (2026-08-10) - Add smooth wheel control for reasoning effort in composer - Show assistant context usage in chat composer - Add Firecrawl as a URL fetch provider for web search - Write a new note straight into the knowledge base - Support scoped AGENTS.md instructions in agent runtime - Add Pi coding agent to Code Tools - Save edited user messages without resending in chat composer - Add Windows Terminal (wt.exe) support for external apps #### v2.0.2 (2026-08-07) - Persist empty branch nodes in chat branch - Built-in web search for DeepSeek, OpenRouter, and Dashscope with distinguishable resolved model names - Add selective cache and legacy data cleanup - Add Claude (Anthropic) conversation import support - Support detailed models through unified gateway in CLI config - Allow rerunning v1 migration in data migration #### v2.0.1 (2026-08-07) - Enable drag reorder for assistant groups - Add 'embedding' endpoint type for NEWAPI model configuration - Keep the create row visible while searching in conversation picker - Support built-in web search for DeepSeek - Show settings as a focused tab in tab bar - Draw a Chrome-style divider between adjacent tabs in tab bar - Read binary office documents through anydoc in knowledge readers #### v2.0.0 (2026-08-05) - Support for both OpenAI and Anthropic endpoints - Built-in core Agent runtime, reducing dependency installation and environment setup - Agents can read and maintain knowledge bases - Multi-window and split-view layouts for multitasking - One-click previews of agent results - One-click model service configuration for Claude Code, Codex, and more in Code Mate ### Changed #### v2.0.12 (2026-09-04) - Markdown previews outside chat now use the same rich renderer as chat messages - Message token details now show time to first token without requiring the expanded runtime breakdown - Sent image attachments now appear as compact filename chips with real image thumbnails, including after reopening a conversation #### v2.0.11 (2026-09-03) - Single-direction text translation now requires only a target language and starts without source-language detection - Confirm screenshot capture on double-click inside the selection rectangle - WebDAV backups now verify the server's TLS certificate by default with option to enable Allow Self-Signed Certificates - Web search now prefers model-native search by default, with automatic fallback to configured services - Improved PDF previews with selectable text, direct page jumps, and document outlines #### v2.0.10 (2026-08-28) - Parallelize independent config pipelines for improved performance - Parallelize device metadata reads for backup performance - Stabilize system prompt prefix in agent - Pace markdown renders and move one-shot highlights to the worker #### v2.0.9 (2026-08-24) - Unify declarative tool approval policy in agent runtime - Lazy-load disabled message outline in chat for performance #### v2.0.5 (2026-08-13) - Reduce cold idle memory with lazy loading - Isolate log tab updates in MCP settings - Consolidate first-paint lazy loading in renderer #### v2.0.4 (2026-08-13) - Serve the trace viewer incrementally and bound its retention - Virtualize large span trees in trace viewer - Route only PDFs through document processors in knowledge readers #### v2.0.3 (2026-08-10) - Eliminate tab-switch IPC and re-render churn under Activity keep-alive in chat tabs #### v2.0.2 (2026-08-07) - Skip redundant re-show work in message tools, artifact pane, and trace page - Make ProxyService own worker routing policy #### v2.0.1 (2026-08-07) - Reduce renderer loading work in agent session #### v2.0.0 (2026-08-05) - Upgraded interactions and visual design for chats, roles, projects, and tasks - Code tools upgraded to Code Mate - Image generation now dynamically shows available parameters for each model - Data foundation upgraded for responsiveness with extensive history ### Fixed #### v2.0.12 (2026-09-04) - Cherry Cloud models now honor server-declared capabilities and support configured OpenAI Chat Completions routes - Cherry Cloud models remain compatible with older model-list responses - Preserve partial chat responses and show a clear, retryable error when a provider stream ends without a finish reason - Fix middle-click autoscroll scrolling in the chat message list to no longer vibrate or jump - Model health check results no longer linger after leaving the provider page - Allow grep searches in Filesystem MCP to use regular expression syntax supported by ripgrep - Empty knowledge bases now appear in the knowledge list - Prevent fraction denominators from being clipped when exporting conversations as images - Prevent the OpenClaw gateway from failing to start when Cherry Studio uses a socks5 proxy - Web fetch no longer sends localhost, LAN, or locally-unresolvable URLs to Jina Reader when the built-in fetch fails - Pinned topics and agent sessions now keep their text aligned with unpinned items in owner-grouped lists #### v2.0.11 (2026-09-03) - Pinned agent sessions now remain in their agent group and appear before unpinned sessions - Knowledge base and web search citations referenced again in a later turn now render as numbered citations and export correctly - Fixed markdown export with images: oversized embeds are skipped, folder-mode images are copied directly, and broken links are removed - Fixed the app freezing for seconds when removing all models of a provider on a large chat history #### v2.0.10 (2026-08-28) - Repair literal autolinks that swallow emphasis markers in chat markdown - Resolve spell check real-time highlighting and multi-language issues - Polish subagent flow and running status - Register the portable Windows launcher at login - Refine skill icon styling - Anchor relative dates to runtime in web search - Prevent nested inputs from starting keyboard drag in sortable - Preserve rapid message navigation targets in chat - Clean up orphaned tab pool state in mini-app - Expose cached tokens in chat completions API - Track app launch once per startup in analytics - Route MiniMax M3 agents through Anthropic Messages - Block image reads for text-only models in claude-code - Tolerate missing annotations in OpenAI Responses replies - Pad composer inset inside code surfaces in artifact pane #### v2.0.9 (2026-08-24) - Support valid GitHub SKILL.md links in skills - Send enable_thinking for Qwen on openai-compatible providers - Prevent pinning transient mini-apps in tabs - Resolve citations for third-party MCP tool outputs in pi-agent - Skip Anthropic tools without an input_schema in api-gateway - Only block skills on provable dependency gaps and split SkillService - Keep Agent SDK system messages out of the assistant turn in api-gateway - Highlight the first frame without debounce delay in code viewer - Drop unanswered approvals from replayed history in ai-messages - Cancel stale requests on conversation switch in message-locate - Recover deleted active topics in topic-navigation - Improve option legibility in permission mode - Ignore stale provider selections in code-mate #### v2.0.8 (2026-08-19) - bind send_message anchors by their target subagent id in dsh-runtime - honor Cherry proxy environment in pi-runtime - terminate parse worker on terminal states in xlsx-preview - add horizontal padding and enable wrapping for text previews - keep macOS title-bar left inset when detached window is fullscreen - skip think:true for Ollama when SDK default triggers auto reasoning - dispatch image + DashScope routing by preset, not raw provider id - parallelize and deduplicate asset loading in shiki - skip irrecoverable Dexie records in data-migration - skip unavailable virtual drive workspaces in agent-migration - clear stale Agent attachment refs on retry in migration - upgrade ONNX Runtime worker lifecycle in local-inference - quit when closing without tray in main-window - keep the styles entry in the bundle so formula copy works in ui-markdown - filter DSH-incompatible providers from Code Mate UI - block unreadable images for non-vision models in attachment-routing - preserve default thinking mode in ollama - use lightweight probe for Ollama health check - inherit Zhipu model provider keys in web-search - prevent duplicate export on double click in obsidian-export - use current Zhipu provider key in web-search - restore Nano Banana 2 parameters in gemini-image - tombstone removed servers to stop ghost stdio clients in mcp-runtime #### v2.0.7 (2026-08-17) - Prevent cross-size model matches in provider registry - Keep code fences when saving a conversation to a knowledge base - Skip in-flight items in bulk reindex for knowledge data source - Recreate session when deleting an agent's only non-active session - Remember the usage page's view selections - Restore 0.01 similarity-threshold precision in knowledge RAG - Stabilize group row height in virtual list for provider model list - Name skills by directory in the injected prompt for agent composer - Clear remembered session id on delete-recovery to stop white-screen crash - Disable reasoning for language detection in translate - Preserve template carousel across tabs in paintings - Re-attach tab to a rebuilt main window in sub-window - Cap MiniMax-M3 output tokens in provider registry - Stop blocking images sent to non-vision models in chat attachments - Preserve soft line breaks in chat messages - Preserve disabled-send feedback in deferred composer fallback #### v2.0.5 (2026-08-13) - Resolve prefixed size-colliding model ids to their own override - Synchronize pin changes across windows - Make cross-assistant topic moves atomic - Bound background memory usage from tab retention - Delete conversations past SQLite's cascade depth limit - Bound inactive conversation history cache - Avoid duplicate refresh in workspace selector - Avoid duplicate tree refetch in topic branch - Virtualize message previews in global search - Query OVMS for its servable status on /v1/config - Prefer chat endpoints for mixed model health checks - Accept parameterized base64 data URLs in file handling - Place newly created agents first in agent list - Register qwen-image-3.0 image models for DashScope - Preserve currency and derive tier input totals in model pricing - Skip invalid legacy model IDs in provider model migration - Validate machine translations before writing them - Preserve chat composer width across topic switches #### v2.0.4 (2026-08-13) - Sync tray-dependent preferences in settings - Propagate stream abort signal into in-flight MCP tool calls - Normalize legacy v1 MCP tool names before model replay - Block unreadable images for text-only models in chat attachments - Find deeply nested workspace file mentions in agent - Keep window shortcuts local in shortcut service - Preserve auto-backup intervals across restarts - Persist assistant group collapse in resource list - Wrap long unbroken lines in wrapped mode instead of clipping in code viewer - Honor auto-collapse setting when disabled for thinking blocks #### v2.0.3 (2026-08-10) - Add sidebar management menu entry - Strip citation markers from IM replies in agent channels - Limit automatic naming to initial turn in agent session - Make the configured system prompt authoritative in agent - Preserve managed files for session history in agent attachments - Respect Linux default terminal in code CLI - Keep input compact in maximized panels in composer - Ship GLIBC-compatible better-sqlite3 for Linux packaging - Always echo reasoning_content on assistant turns for OpenAI-compatible - Dismiss loading overlay on dom-ready in mini app - Skip EBUSY locked files during automatic backup - Recover duplicate tool use ID failures in agent runtime #### v2.0.2 (2026-08-07) - Reap SDK CLI subprocesses on shutdown in claude-code - Raise the default tool-call round cap - Close the snapshot-to-stream delivery gap in directory-tree - Allow direct upgrades to v2.0.x in v2 migration - Open image tokens in shared preview in composer - Start WeChat QR login in agent channels - Preserve form edits and improve tool details in MCP settings - Serve every endpoint from one host with per-route versions in new API - Skip symlinks during migration in agent migration - Survive mirror failures and keep embedding inference offline in local model - Respect Traditional Chinese locale in selection assistant #### v2.0.1 (2026-08-07) - Localize workspace tree errors in agent files - Use preset append mode for agent prompts in claude-code - Serve DeepSeek V4 models over the Anthropic endpoint - Recover stale dashboards across windows in openclaw - Prevent reference context leaks in collapsed previews - Normalize no-prefill agent requests in API gateway - Persist provider filter in provider settings - Hide prompt variable hint in agent dialog - Use free Qwen instead of the default assistant model in AI diagnosis - Allow WebP avatar uploads in profile - Isolate nested scroll regions from chat viewport - Hide disabled MCP tools from agents in claude-code - Preserve TeX when copying formulas in selection menu - Add bidirectional controls to horizontal scroll container - Show model identity on single replies in chat - Avoid redundant first-page requests in resource pagination - Normalize internal Agent continuations in API gateway - Bypass proxy for loopback gateways in agent runtime - Align export and save actions in export menu - Strip schema dialect metadata in gemini tools - Add model type scroll controls in provider settings - Tolerate unusable legacy API keys in provider migration - Restore default reasoning effort in composer - Restore normal agent capabilities in cherry assistant - Preserve conversation tab identity in tab navigation - Restore assistant identity for migrated messages in chat - Preserve lazy-loaded data between openings in model selector - Support Claude tools in Responses API in API gateway - Guard query type in WebSearchTool and normalizeWebSearchKeywords, fix translation button contrast - Prevent crash by adding error handling to V2 migration pipeline - Stop repeated saves after failures in resource catalog - Prevent filtered model selection crash in model selector - Migrate tokenflux painting history in painting migrator - Route OpenCode Go models by their served protocol in provider registry #### v2.0.0 (2026-08-05) - Improved scrolling performance in very long conversations - Large-folder imports no longer block the interface ### Removed #### v2.0.7 (2026-08-17) - Retire the per-agent max_turns cap in agent configuration ### Security #### v2.0.8 (2026-08-19) - close four bypass gaps in existing guards including scheme checks, staging residue, and trace perms #### v2.0.1 (2026-08-07) - Address high and critical security advisories ## Release notes ### v2.0.12 - Date: 2026-09-04 - Version: v2.0.12 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.12 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.12 - **added** — Settings gains a sidebar search to find any setting row across all sections including title, description, aliases, and Chinese pinyin, with scroll, highlight, and full keyboard navigation - **added** — Settings search now covers all sections with static settings rows, including MCP tabs - **added** — Translate model now has its own temperature, top-p, and reasoning effort settings, editable from the translation settings panel and translate page toolbar - **changed** — Markdown previews outside chat now use the same rich renderer as chat messages - **added** — In agent session messages and rendered artifact Markdown previews, workspace links open files in the artifact pane and directories in the system file manager, with relative links resolving from the agent workspace root - **fixed** — Cherry Cloud models now honor server-declared capabilities and support configured OpenAI Chat Completions routes - **fixed** — Cherry Cloud models remain compatible with older model-list responses - **changed** — Message token details now show time to first token without requiring the expanded runtime breakdown - **fixed** — Preserve partial chat responses and show a clear, retryable error when a provider stream ends without a finish reason - **fixed** — Fix middle-click autoscroll scrolling in the chat message list to no longer vibrate or jump - **changed** — Sent image attachments now appear as compact filename chips with real image thumbnails, including after reopening a conversation - **fixed** — Model health check results no longer linger after leaving the provider page - **fixed** — Allow grep searches in Filesystem MCP to use regular expression syntax supported by ripgrep - **fixed** — Empty knowledge bases now appear in the knowledge list - **fixed** — Prevent fraction denominators from being clipped when exporting conversations as images - **fixed** — Prevent the OpenClaw gateway from failing to start when Cherry Studio uses a socks5 proxy - **fixed** — Web fetch no longer sends localhost, LAN, or locally-unresolvable URLs to Jina Reader when the built-in fetch fails - **fixed** — Pinned topics and agent sessions now keep their text aligned with unpinned items in owner-grouped lists ##### Downloads / 下载 (v2.0.12) | Platform | Architecture | Global | China Edition | | --- | --- | --- | --- | | Windows | x64 | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-win-x64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-win-x64-portable.exe) | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-win-x64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-win-x64-portable.exe) | | Windows | arm64 | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-win-arm64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-win-arm64-portable.exe) | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-win-arm64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-win-arm64-portable.exe) | | macOS | x64 | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-mac-x64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-mac-x64.zip) | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-mac-x64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-mac-x64.zip) | | macOS | arm64 | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-mac-arm64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-mac-arm64.zip) | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-mac-arm64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-mac-arm64.zip) | | Linux | x64 | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-linux-x64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-linux-x64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-linux-x64.rpm) | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-linux-x64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-linux-x64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-linux-x64.rpm) | | Linux | arm64 | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-linux-arm64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-linux-arm64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-2.0.12-linux-arm64.rpm) | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-linux-arm64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-linux-arm64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.12/Cherry-Studio-CN-2.0.12-linux-arm64.rpm) | ##### Release Notes / 发布说明
English Cherry Studio 2.0.12 - Features, Fixes, and Improvements ✨ New Features - [Settings] Settings gains a sidebar search: find any setting row across all sections (title/description/aliases, Chinese pinyin included), jump to it with scroll + highlight, full keyboard navigation. - [Settings] Settings search now covers all sections with static settings rows, including MCP tabs. - _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.12]_ ### v2.0.11 - Date: 2026-09-03 - Version: v2.0.11 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.11 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.11 - **added** — Sign in to Cherry Cloud for entitlement-aware Agent models, automatic local Gateway startup, quota-aware model selection, and durable Product Session handling - **added** — Configure the compression trigger threshold per assistant or globally in Settings - **added** — Improved installation and navigation for Mini Apps with drag-and-drop packages, stable tabs and icons, and transparent Translate and Painting views - **changed** — Single-direction text translation now requires only a target language and starts without source-language detection - **added** — Press Ctrl+A (Cmd+A on macOS) in the file list to select all files in the current view - **added** — Added an inline delete button to remove a message's translation in one click - **added** — Hide the chat assistant from the sidebar via Appearance Settings, and launch directly to the first visible sidebar app - **added** — Added a global search action to the Agent conversation toolbar - **added** — Add Hy4 Preview, Ling 3.0 Flash, and Qwen3.8 Flash model support - **added** — Anthropic-compatible providers can now keep prompt caches for 1 hour from API Settings - **added** — Added initial Turkish localization support across Cherry Studio - **changed** — Confirm screenshot capture on double-click inside the selection rectangle - **fixed** — Pinned agent sessions now remain in their agent group and appear before unpinned sessions - **fixed** — Knowledge base and web search citations referenced again in a later turn now render as numbered citations and export correctly - **changed** — WebDAV backups now verify the server's TLS certificate by default with option to enable Allow Self-Signed Certificates - **changed** — Web search now prefers model-native search by default, with automatic fallback to configured services - **fixed** — Fixed markdown export with images: oversized embeds are skipped, folder-mode images are copied directly, and broken links are removed - **added** — Shows the available disk space when a backup or restore fails because the disk is full - **fixed** — Fixed the app freezing for seconds when removing all models of a provider on a large chat history - **changed** — Improved PDF previews with selectable text, direct page jumps, and document outlines ##### Downloads / 下载 (v2.0.11) | Platform | Architecture | Global | China Edition | | --- | --- | --- | --- | | Windows | x64 | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-win-x64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-win-x64-portable.exe) | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-win-x64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-win-x64-portable.exe) | | Windows | arm64 | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-win-arm64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-win-arm64-portable.exe) | [Installer](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-win-arm64-setup.exe) · [Portable](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-win-arm64-portable.exe) | | macOS | x64 | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-mac-x64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-mac-x64.zip) | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-mac-x64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-mac-x64.zip) | | macOS | arm64 | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-mac-arm64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-mac-arm64.zip) | [DMG](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-mac-arm64.dmg) · [ZIP](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-mac-arm64.zip) | | Linux | x64 | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-linux-x64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-linux-x64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-linux-x64.rpm) | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-linux-x64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-linux-x64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-linux-x64.rpm) | | Linux | arm64 | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-linux-arm64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-linux-arm64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-2.0.11-linux-arm64.rpm) | [AppImage](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-linux-arm64.AppImage) · [DEB](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-linux-arm64.deb) · [RPM](https://github.com/CherryHQ/cherry-studio/releases/download/v2.0.11/Cherry-Studio-CN-2.0.11-linux-arm64.rpm) | ##### Release Notes / 发布说明
English Cherry Studio 2.0.11 - Features, Fixes, and Improvements ✨ New Features - [Cherry Cloud] Sign in to Cherry Cloud for entitlement-aware Agent models, automatic local Gateway startup, quota-aware model selection, and durable Product Session handling. - [Context Management] Configure the compression trigger threshold per assistant or globally in Settings. - [Mini Apps] Improve _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.11]_ ### v2.0.10 - Date: 2026-08-28 - Version: v2.0.10 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.10 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.10 - **fixed** — Repair literal autolinks that swallow emphasis markers in chat markdown - **fixed** — Resolve spell check real-time highlighting and multi-language issues - **fixed** — Polish subagent flow and running status - **changed** — Parallelize independent config pipelines for improved performance - **fixed** — Register the portable Windows launcher at login - **fixed** — Refine skill icon styling - **fixed** — Anchor relative dates to runtime in web search - **fixed** — Prevent nested inputs from starting keyboard drag in sortable - **fixed** — Preserve rapid message navigation targets in chat - **changed** — Parallelize device metadata reads for backup performance - **changed** — Stabilize system prompt prefix in agent - **added** — Show run status on task cards for scheduled tasks - **fixed** — Clean up orphaned tab pool state in mini-app - **fixed** — Expose cached tokens in chat completions API - **fixed** — Track app launch once per startup in analytics - **changed** — Pace markdown renders and move one-shot highlights to the worker - **fixed** — Route MiniMax M3 agents through Anthropic Messages - **fixed** — Block image reads for text-only models in claude-code - **fixed** — Tolerate missing annotations in OpenAI Responses replies - **fixed** — Pad composer inset inside code surfaces in artifact pane ##### What's Changed * fix(chat-markdown): repair literal autolinks that swallow emphasis markers by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/19136 * fix: resolve spell check real-time highlighting and multi-language issues by @LWWZH in https://github.com/CherryHQ/cherry-studio/pull/14634 * fix(agent): polish subagent flow and running status by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/19309 * perf(cli): parallelize independent config pipelines by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19284 * fix(app): register the portable Windows launcher at login by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19285 * fix(skills): refine skill icon styling by @SiinXu in https://github.com/CherryHQ/cherry-studio/pull/18421 * docs(cherry-electron-dev): expand performance diagnostics by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18458 * fix(web-search): anchor relative dates to runtime by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19304 * docs(cherry-electron-dev): fix broken diagnostics link by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/19354 * fix(sortable): prevent nested inputs from starting keyboard drag by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/19344 * fix(chat): preserve rapid message navigation targets by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/19189 * perf(backup): parallelize device metadata reads by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19282 * perf(agent): stabilize system prompt prefix by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19297 * feat(scheduled-tasks): show run status on task cards by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18522 * fix(mini-app): clean up orphaned tab pool state by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/16788 * fix(api): expose cached tokens in chat completions by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19353 * fix(analytics): track app launch once per startup by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/16670 * perf(translate,shiki): pace markdown renders and move one-shot highlights to the worker by @GeorgeDong32 in https://github.com/CherryHQ/cherry-studio/pull/18926 * fix(provider-registry): route MiniMax M3 agents through Anthropic Messages by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/19361 * fix(claude-code): block image reads for text-only models by @iris-Neko in https://github.com/CherryHQ/cherry-studio/pull/19366 * fix(ai-provider): tolerate missing annotations in OpenAI Responses replies by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/19365 * fix(artifact-pane): pad composer inset inside code surfaces by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/19372 * fix(selection-assistant): keep custom action footer visible by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19373 * fix(composer): decouple link chip color from user theme by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/19378 * fix(analytics): update SDK and filter token usage reports by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/19352 * fix(mini-app): realign tab pool current state from consistent snapshots by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/19375 * fix(google-provider): send includeServerSideToolInvocations conditionally for Gemini 3 tool combination by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/19364 * feat(model-drawer): add token limit presets by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/19384 * fix(binary-manager): repair incomplete DSH installs by @Totoro-qaq in https://github.com/CherryHQ/cherry-studio/pull/19374 * fix(assistant-settings): name custom parameter value inputs by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19168 * fix(quick-panel): _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.10]_ ### v2.0.9 - Date: 2026-08-24 - Version: v2.0.9 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.9 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.9 - **fixed** — Support valid GitHub SKILL.md links in skills - **added** — Enable remote catalog hot-update without an app release in provider registry - **changed** — Unify declarative tool approval policy in agent runtime - **fixed** — Send enable_thinking for Qwen on openai-compatible providers - **fixed** — Prevent pinning transient mini-apps in tabs - **changed** — Lazy-load disabled message outline in chat for performance - **fixed** — Resolve citations for third-party MCP tool outputs in pi-agent - **fixed** — Skip Anthropic tools without an input_schema in api-gateway - **added** — Publish August 2026 privacy policy update - **added** — Restore Ctrl+L clear messages shortcut - **fixed** — Only block skills on provable dependency gaps and split SkillService - **fixed** — Keep Agent SDK system messages out of the assistant turn in api-gateway - **added** — Add model settings navigation in assistant settings - **fixed** — Highlight the first frame without debounce delay in code viewer - **fixed** — Drop unanswered approvals from replayed history in ai-messages - **fixed** — Cancel stale requests on conversation switch in message-locate - **fixed** — Recover deleted active topics in topic-navigation - **fixed** — Improve option legibility in permission mode - **fixed** — Ignore stale provider selections in code-mate - **added** — Pin assistants and agents to the sidebar ##### What's Changed * fix(skills): support valid GitHub SKILL.md links by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18907 * feat(provider-registry): remote catalog hot-update without an app release by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/16857 * refactor(agent-runtime): unify declarative tool approval policy by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18865 * fix(provider-registry): send enable_thinking for Qwen on openai-compatible providers by @404-Page-Found in https://github.com/CherryHQ/cherry-studio/pull/18772 * fix(tabs): prevent pinning transient mini-apps by @userInner in https://github.com/CherryHQ/cherry-studio/pull/18904 * perf(chat): lazy-load disabled message outline by @ActiveSky in https://github.com/CherryHQ/cherry-studio/pull/18971 * fix(pi-agent): resolve citations for third-party MCP tool outputs by @404-Page-Found in https://github.com/CherryHQ/cherry-studio/pull/18829 * fix(api-gateway): skip Anthropic tools without an input_schema by @userInner in https://github.com/CherryHQ/cherry-studio/pull/18780 * docs(code-review): require dual sources for network downloads by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18953 * feat(privacy): publish August 2026 policy update by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/19022 * ci(claude-review): keep review failures advisory by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/19031 * feat(shortcuts): restore Ctrl+L clear messages shortcut by @SiinXu in https://github.com/CherryHQ/cherry-studio/pull/19021 * fix(skills): only block skills on provable dependency gaps, and split SkillService by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18549 * fix(api-gateway): keep Agent SDK system messages out of the assistant turn by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18697 * feat(assistant-settings): add model settings navigation by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/19030 * fix(code-viewer): highlight the first frame without the debounce delay by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18950 * fix(ai-messages): drop unanswered approvals from replayed history by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18066 * fix(message-locate): cancel stale requests on conversation switch by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18994 * fix(topic-navigation): recover deleted active topics by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18995 * fix(permission-mode): improve option legibility by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/19035 * fix(code-mate): ignore stale provider selections by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/19015 * fix(job-scheduler): persist automatic next runs by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/19017 * fix(api-gateway): downgrade inline system messages for Ollama by @userInner in https://github.com/CherryHQ/cherry-studio/pull/19042 * refactor(mcp): move builtin server knowledge out of the MCP runtime by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18630 * fix(pdf-translation): show resource download progress by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/19051 * fix(migration): preserve agent identity files when v1 source workspace is removed by @404-Page-Found in https://github.com/CherryHQ/cherry-studio/pull/19053 * refactor(external-app): unify external open targets by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/19027 * feat(local-model): enable inference hardware acceleration by default by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/19055 * test(job-scheduler): fix flaky nextRun assertion on natural interval fire by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/19057 * fix(agent-session): update lastActivityAt when reusing _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.9]_ ### v2.0.8 - Date: 2026-08-19 - Version: v2.0.8 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.8 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.8 - **fixed** — bind send_message anchors by their target subagent id in dsh-runtime - **fixed** — honor Cherry proxy environment in pi-runtime - **added** — add skill global enablement controls - **added** — add floating task progress capsule to agents - **fixed** — terminate parse worker on terminal states in xlsx-preview - **fixed** — add horizontal padding and enable wrapping for text previews - **fixed** — keep macOS title-bar left inset when detached window is fullscreen - **fixed** — skip think:true for Ollama when SDK default triggers auto reasoning - **added** — add interactive screen capture with annotation and OCR - **fixed** — dispatch image + DashScope routing by preset, not raw provider id - **fixed** — parallelize and deduplicate asset loading in shiki - **fixed** — skip irrecoverable Dexie records in data-migration - **fixed** — skip unavailable virtual drive workspaces in agent-migration - **fixed** — clear stale Agent attachment refs on retry in migration - **fixed** — upgrade ONNX Runtime worker lifecycle in local-inference - **fixed** — quit when closing without tray in main-window - **fixed** — keep the styles entry in the bundle so formula copy works in ui-markdown - **fixed** — filter DSH-incompatible providers from Code Mate UI - **fixed** — block unreadable images for non-vision models in attachment-routing - **fixed** — preserve default thinking mode in ollama - **added** — add TypeScript-guided code mode tools to pi-runtime - **added** — add layout-preserving PDF translation via BabelDOC - **added** — record translations in history and file manager for pdf-translation - **added** — surface MCP prompts and resources in the composer and to the model - **fixed** — use lightweight probe for Ollama health check - **fixed** — inherit Zhipu model provider keys in web-search - **fixed** — prevent duplicate export on double click in obsidian-export - **fixed** — use current Zhipu provider key in web-search - **fixed** — restore Nano Banana 2 parameters in gemini-image - **fixed** — tombstone removed servers to stop ghost stdio clients in mcp-runtime - **added** — add durable cross-session delivery to agent-session - **added** — allow private-network fetch targets behind a preference in remote-fetch - **security** — close four bypass gaps in existing guards including scheme checks, staging residue, and trace perms ##### What's Changed * fix(dsh-runtime): bind send_message anchors by their target subagent id by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18716 * fix(pi-runtime): honor Cherry proxy environment by @userInner in https://github.com/CherryHQ/cherry-studio/pull/18778 * feat(skills): add skill global enablement controls by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18185 * feat(agents): add floating task progress capsule by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18469 * fix(xlsx-preview): terminate parse worker on terminal states by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18773 * fix(file-preview): add horizontal padding and enable wrapping for text previews by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18752 * fix(sub-window): keep macOS title-bar left inset when detached window is fullscreen by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18398 * fix(api-gateway): skip think:true for Ollama when SDK default triggers auto reasoning by @404-Page-Found in https://github.com/CherryHQ/cherry-studio/pull/18744 * feat(screenshot): add interactive screen capture with annotation and OCR by @0xfullex in https://github.com/CherryHQ/cherry-studio/pull/18725 * fix(ai-provider): dispatch image + DashScope routing by preset, not raw provider id by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18735 * fix(shiki): parallelize and deduplicate asset loading by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18627 * ci(i18n): require complete translations in pull requests by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18750 * fix(data-migration): skip irrecoverable Dexie records by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18783 * fix(agent-migration): skip unavailable virtual drive workspaces by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18727 * fix(migration): clear stale Agent attachment refs on retry by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18757 * fix(local-inference): upgrade ONNX Runtime worker lifecycle by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18785 * fix(main-window): quit when closing without tray by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18448 * fix(ui-markdown): keep the styles entry in the bundle so formula copy works by @he-yufeng in https://github.com/CherryHQ/cherry-studio/pull/18711 * fix(code-mate): filter DSH-incompatible providers from Code Mate UI by @404-Page-Found in https://github.com/CherryHQ/cherry-studio/pull/18771 * fix(attachment-routing): block unreadable images for non-vision models by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18797 * fix(ollama): preserve default thinking mode by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18784 * fix(claude-review): rerun reviews after PR updates by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18805 * feat(pi-runtime): add TypeScript-guided code mode tools by @vaayne in https://github.com/CherryHQ/cherry-studio/pull/18708 * refactor(settings): reorganize general preferences by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18794 * fix(i18n): add Code Mode tool translations by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18812 * docs(claude-md): scope pre-commit checks to the change at hand by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18808 * fix(chat-message): restore editing for provider-annotated assistant replies by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18635 * fix(image-generation): retry image requests without response_format on 400 by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18702 * fix(code-mate): default unconfigured launches to unified gateway by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18811 * feat(pdf-translation): layout-preserv _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.8]_ ### v2.0.7 - Date: 2026-08-17 - Version: v2.0.7 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.7 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.7 - **fixed** — Prevent cross-size model matches in provider registry - **added** — Add sidebar help and diagnostic upload to help-feedback - **fixed** — Keep code fences when saving a conversation to a knowledge base - **fixed** — Skip in-flight items in bulk reindex for knowledge data source - **fixed** — Recreate session when deleting an agent's only non-active session - **removed** — Retire the per-agent max_turns cap in agent configuration - **fixed** — Remember the usage page's view selections - **fixed** — Restore 0.01 similarity-threshold precision in knowledge RAG - **fixed** — Stabilize group row height in virtual list for provider model list - **fixed** — Name skills by directory in the injected prompt for agent composer - **added** — Add auto permission mode for pi agents and flatten the runtime picker - **fixed** — Clear remembered session id on delete-recovery to stop white-screen crash - **fixed** — Disable reasoning for language detection in translate - **added** — Integrate DeepSeek Harness in code-mate - **fixed** — Preserve template carousel across tabs in paintings - **fixed** — Re-attach tab to a rebuilt main window in sub-window - **fixed** — Cap MiniMax-M3 output tokens in provider registry - **fixed** — Stop blocking images sent to non-vision models in chat attachments - **fixed** — Preserve soft line breaks in chat messages - **fixed** — Preserve disabled-send feedback in deferred composer fallback ##### What's Changed * fix(provider-registry): prevent cross-size model matches by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18513 * feat(help-feedback): add sidebar help and diagnostic upload by @SiinXu in https://github.com/CherryHQ/cherry-studio/pull/18198 * docs: remove obsolete v1-maintenance constraints and stale v2 wording by @GeorgeDong32 in https://github.com/CherryHQ/cherry-studio/pull/18548 * fix(knowledge-content): keep code fences when saving a conversation to a knowledge base by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18567 * fix(knowledge-data-source): skip in-flight items in bulk reindex by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18578 * fix(agents): recreate session when deleting an agent's only non-active session by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18576 * feat(agent-config): retire the per-agent max_turns cap by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18579 * fix(usage-settings): remember the usage page's view selections by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18575 * fix(knowledge-rag): restore 0.01 similarity-threshold precision by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18566 * fix(provider-model-list): stabilize group row height in virtual list by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18568 * fix(agent-composer): name skills by directory in the injected prompt by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18561 * feat(agent-runtime): add auto permission mode for pi agents and flatten the runtime picker by @vaayne in https://github.com/CherryHQ/cherry-studio/pull/18551 * fix(agent-session): clear remembered session id on delete-recovery to stop white-screen crash by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18577 * fix(translate): disable reasoning for language detection by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18574 * feat(code-mate): integrate DeepSeek Harness by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18565 * fix(paintings): preserve template carousel across tabs by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18538 * fix(sub-window): re-attach tab to a rebuilt main window by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18285 * fix(provider-registry): cap MiniMax-M3 output tokens by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18591 * fix(chat-attachments): stop blocking images sent to "non-vision" models by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18582 * fix(chat-markdown): preserve soft line breaks in chat messages by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18530 * fix(composer): preserve disabled-send feedback in deferred composer fallback by @LWWZH in https://github.com/CherryHQ/cherry-studio/pull/18602 * fix(ui): dismiss action popovers after selection by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18605 * refactor(composer): lift agent path containment into renderer path utils by @EurFelux in https://github.com/CherryHQ/cherry-studio/pull/17718 * fix(topics): clear unlinked assistant group when its last topic is deleted by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18446 * fix(window-manager): re-assert alwaysOnTop on every platform after show by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18641 * fix(agent-session): switch to neighbour before deleting active session by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18570 * fix(composer): collapse selection after clearing input by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18649 * fix(mcp): redact env secrets from serverKey and builtin server logs by @GeorgeDong32 in https://github.com/CherryHQ/cherry-stu _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.7]_ ### v2.0.5 - Date: 2026-08-13 - Version: v2.0.5 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.5 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.5 - **fixed** — Resolve prefixed size-colliding model ids to their own override - **changed** — Reduce cold idle memory with lazy loading - **fixed** — Synchronize pin changes across windows - **fixed** — Make cross-assistant topic moves atomic - **fixed** — Bound background memory usage from tab retention - **fixed** — Delete conversations past SQLite's cascade depth limit - **changed** — Isolate log tab updates in MCP settings - **fixed** — Bound inactive conversation history cache - **added** — Add built-in support agent - **fixed** — Avoid duplicate refresh in workspace selector - **fixed** — Avoid duplicate tree refetch in topic branch - **fixed** — Virtualize message previews in global search - **added** — Add app update notification preference - **fixed** — Query OVMS for its servable status on /v1/config - **fixed** — Prefer chat endpoints for mixed model health checks - **fixed** — Accept parameterized base64 data URLs in file handling - **changed** — Consolidate first-paint lazy loading in renderer - **fixed** — Place newly created agents first in agent list - **fixed** — Register qwen-image-3.0 image models for DashScope - **fixed** — Preserve currency and derive tier input totals in model pricing - **added** — Install a skill from a GitHub SKILL.md URL - **fixed** — Skip invalid legacy model IDs in provider model migration - **fixed** — Validate machine translations before writing them - **fixed** — Preserve chat composer width across topic switches ##### What's Changed * fix(provider-registry): resolve prefixed size-colliding model ids to their own override by @404-Page-Found in https://github.com/CherryHQ/cherry-studio/pull/18487 * perf(main-runtime): reduce cold idle memory with lazy loading by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18472 * fix(pins): synchronize pin changes across windows by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18499 * fix(topic-move): make cross-assistant moves atomic by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18501 * fix(tab-retention): bound background memory usage by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18504 * fix(message-service): delete conversations past SQLite's cascade depth limit by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18479 * perf(mcp-settings): isolate log tab updates by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18473 * fix(data-api): bound inactive conversation history cache by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18505 * ci(update-mergeable-prs): merge main into approved PRs that are behind by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18524 * feat(cherry-support): add built-in support agent by @SiinXu in https://github.com/CherryHQ/cherry-studio/pull/18498 * fix(workspace-selector): avoid duplicate refresh by @KXHXK in https://github.com/CherryHQ/cherry-studio/pull/18475 * fix(topic-branch): avoid duplicate tree refetch by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18470 * fix(global-search): virtualize message previews by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18402 * feat(notification): add app update notification preference by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18332 * fix(model-list): query OVMS for its servable status on /v1/config by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18482 * fix: prefer chat endpoints for mixed model health checks by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18447 * fix(file): accept parameterized base64 data URLs by @VectorPeak in https://github.com/CherryHQ/cherry-studio/pull/17406 * perf(renderer): consolidate first-paint lazy loading by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18405 * fix(agent-list): place newly created agents first by @SiinXu in https://github.com/CherryHQ/cherry-studio/pull/18427 * fix(dashscope): register qwen-image-3.0 image models by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18527 * fix(model-pricing): preserve currency and derive tier input totals by @chujiamo in https://github.com/CherryHQ/cherry-studio/pull/18453 * feat(skill-install): install a skill from a GitHub SKILL.md URL by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18438 * fix(provider-model-migration): skip invalid legacy model IDs by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18381 * fix(i18n-translate): validate machine translations before writing them by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18510 * fix(chat-composer): preserve width across topic switches by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18529 * fix(agent-migration): continue past overlapping workspaces by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18372 * fix(file-migration): copy trailing-space/dot ext uploads to their v2 name by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18514 * fix(topic-naming): use quick assistant model by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18517 * fix(chat-attachments): sanitize bad mediaType before provider dispatch by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18481 * fix(conversation-sidebar): preserve collapsed pane state by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18507 * fix(artifact-pane): _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.5]_ ### v2.0.4 - Date: 2026-08-13 - Version: v2.0.4 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.4 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.4 - **added** — Add back-to-top button and line-break mode to notes - **added** — Add delete impact preview to agent workspace - **added** — Add tab navigation shortcuts - **added** — Restore MCP-over-HTTP endpoints (/v1/mcps) in API gateway - **added** — Add opt-in MCP sessions with server-to-client push and progress in API gateway - **added** — Handle GROUP_MESSAGE_CREATE event with per-channel mention-only toggle for QQ channel - **changed** — Serve the trace viewer incrementally and bound its retention - **changed** — Virtualize large span trees in trace viewer - **changed** — Route only PDFs through document processors in knowledge readers - **fixed** — Sync tray-dependent preferences in settings - **fixed** — Propagate stream abort signal into in-flight MCP tool calls - **fixed** — Normalize legacy v1 MCP tool names before model replay - **fixed** — Block unreadable images for text-only models in chat attachments - **fixed** — Find deeply nested workspace file mentions in agent - **fixed** — Keep window shortcuts local in shortcut service - **fixed** — Preserve auto-backup intervals across restarts - **fixed** — Persist assistant group collapse in resource list - **fixed** — Wrap long unbroken lines in wrapped mode instead of clipping in code viewer - **fixed** — Honor auto-collapse setting when disabled for thinking blocks ##### What's Changed * test(ai): add chat turn integration trajectories by @xjt-box in https://github.com/CherryHQ/cherry-studio/pull/17550 * perf(ai-trace): serve the trace viewer incrementally and bound its retention by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18203 * ci(prepare-release): sync and translate i18n before cutting a release by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18214 * fix(settings): sync tray-dependent preferences by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/17445 * feat(notes): add back-to-top button and line-break mode by @LWWZH in https://github.com/CherryHQ/cherry-studio/pull/17788 * feat(agent-workspace): add delete impact preview by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18164 * fix(mcp-runtime): propagate stream abort signal into in-flight MCP tool calls by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18294 * fix(ai-messages): normalize legacy v1 MCP tool names before model replay by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18224 * fix(chat-attachments): block unreadable images for text-only models by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18297 * refactor(knowledge-readers): route only PDFs through document processors by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18161 * fix(agent): find deeply nested workspace file mentions by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18139 * fix(shortcut-service): keep window shortcuts local by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/18282 * fix(backup): preserve auto-backup intervals across restarts by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18226 * fix(resource-list): persist assistant group collapse by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18252 * docs(claude-md): cap inline comments, ban behavior-pinning tests, drop linear-history rule by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18302 * fix(code-viewer): wrap long unbroken lines in wrapped mode instead of clipping by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18290 * fix(thinking-block): honor auto-collapse setting when disabled by @linktao159357 in https://github.com/CherryHQ/cherry-studio/pull/18073 * fix(topic-naming): keep auto-rename requests free of assistant tool config by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18320 * fix(mini-app): reuse existing sidebar tab by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18233 * fix(assistant-editor): recover from legacy save failures by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/18257 * perf(trace-viewer): virtualize large span trees by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18296 * fix(knowledge): add cursor pagination to knowledge base lists by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18300 * fix(system-ocr): send image bytes on Windows so JPEG decodes by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18335 * refactor(file-utils): split path spec from filename convention, add RelativeFilePath brands by @EurFelux in https://github.com/CherryHQ/cherry-studio/pull/18229 * refactor(knowledge): type material paths as PosixRelativeFilePath by @EurFelux in https://github.com/CherryHQ/cherry-studio/pull/18230 * feat(shortcuts): add tab navigation shortcuts by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/16629 * fix(windows-env): preserve Unicode command paths by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18201 * fix(dashscope): skip web_extractor when thinking is disabled by @tommyzhang100504 in https://github.com/CherryHQ/cherry-studio/pull/18349 * fix(agent-editor): let an explicit close discard a failed agent save by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/18331 * fix(compose _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.4]_ ### v2.0.3 - Date: 2026-08-10 - Version: v2.0.3 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.3 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.3 - **added** — Add smooth wheel control for reasoning effort in composer - **added** — Show assistant context usage in chat composer - **added** — Add Firecrawl as a URL fetch provider for web search - **added** — Write a new note straight into the knowledge base - **added** — Support scoped AGENTS.md instructions in agent runtime - **added** — Add Pi coding agent to Code Tools - **added** — Save edited user messages without resending in chat composer - **added** — Add Windows Terminal (wt.exe) support for external apps - **fixed** — Add sidebar management menu entry - **fixed** — Strip citation markers from IM replies in agent channels - **fixed** — Limit automatic naming to initial turn in agent session - **fixed** — Make the configured system prompt authoritative in agent - **fixed** — Preserve managed files for session history in agent attachments - **fixed** — Respect Linux default terminal in code CLI - **fixed** — Keep input compact in maximized panels in composer - **fixed** — Ship GLIBC-compatible better-sqlite3 for Linux packaging - **fixed** — Always echo reasoning_content on assistant turns for OpenAI-compatible - **fixed** — Dismiss loading overlay on dom-ready in mini app - **fixed** — Skip EBUSY locked files during automatic backup - **fixed** — Recover duplicate tool use ID failures in agent runtime - **changed** — Eliminate tab-switch IPC and re-render churn under Activity keep-alive in chat tabs ##### What's Changed * feat(composer): add smooth wheel control for reasoning effort by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/17829 * fix(sidebar): add sidebar management menu entry by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18031 * fix(agent-channels): strip citation markers from IM replies by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/18064 * feat(chat-composer): show assistant context usage by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18006 * fix(agent-session): limit automatic naming to initial turn by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17987 * feat(web-search): add Firecrawl as a URL fetch provider by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18089 * perf(chat-tabs): eliminate tab-switch IPC and re-render churn under Activity keep-alive by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18035 * feat(knowledge-data-source): write a new note straight into the knowledge base by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18022 * fix(agent): make the configured system prompt authoritative by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18100 * fix(agent-attachments): preserve managed files for session history by @vaayne in https://github.com/CherryHQ/cherry-studio/pull/18027 * fix(code-cli): respect Linux default terminal by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/18103 * fix(composer): keep input compact in maximized panels by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18125 * docs(design-system): streamline global UI guidance by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18110 * fix(linux-packaging): ship GLIBC-compatible better-sqlite3 by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/17778 * fix(openai-compatible): always echo reasoning_content on assistant turns by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18121 * fix(mini-app): dismiss loading overlay on dom-ready by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/18079 * feat(agent-runtime): support scoped AGENTS.md instructions by @vaayne in https://github.com/CherryHQ/cherry-studio/pull/18126 * fix(backup): skip EBUSY locked files (LevelDB LOCK) during automatic backup by @xxiaoxiong in https://github.com/CherryHQ/cherry-studio/pull/18117 * fix(agent-runtime): recover duplicate tool use ID failures by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18119 * fix(mcp): resolve stdio MCP servers to full paths on Windows by @LWWZH in https://github.com/CherryHQ/cherry-studio/pull/16472 * fix(notes): align sidebar note icons by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/18143 * fix(mcp): close stale clients so ping failures stop orphaning child processes by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18146 * fix(chat-migration): bound message block indexing memory by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18127 * ci: reduce pre-merge test runtime by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/18165 * fix(lifecycle): bound each service's teardown during shutdown by @0xfullex in https://github.com/CherryHQ/cherry-studio/pull/18136 * feat(code-cli): add Pi coding agent to Code Tools by @vaayne in https://github.com/CherryHQ/cherry-studio/pull/16928 * fix(mcp-tools): prevent non-ASCII tool identity collisions by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18128 * fix(builtin-tools): stop sending strict so knowledge bases work on Anthropic by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18179 * fix(backup): skip LevelDB locks in streamed backups by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18173 * fix(resource-list): unify cross-panel list styling by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18046 * fi _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.3]_ ### v2.0.2 - Date: 2026-08-07 - Version: v2.0.2 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.2 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.2 - **added** — Persist empty branch nodes in chat branch - **changed** — Skip redundant re-show work in message tools, artifact pane, and trace page - **fixed** — Reap SDK CLI subprocesses on shutdown in claude-code - **fixed** — Raise the default tool-call round cap - **fixed** — Close the snapshot-to-stream delivery gap in directory-tree - **added** — Built-in web search for DeepSeek, OpenRouter, and Dashscope with distinguishable resolved model names - **fixed** — Allow direct upgrades to v2.0.x in v2 migration - **added** — Add selective cache and legacy data cleanup - **fixed** — Open image tokens in shared preview in composer - **fixed** — Start WeChat QR login in agent channels - **changed** — Make ProxyService own worker routing policy - **added** — Add Claude (Anthropic) conversation import support - **added** — Support detailed models through unified gateway in CLI config - **added** — Allow rerunning v1 migration in data migration - **fixed** — Preserve form edits and improve tool details in MCP settings - **fixed** — Serve every endpoint from one host with per-route versions in new API - **fixed** — Skip symlinks during migration in agent migration - **fixed** — Survive mirror failures and keep embedding inference offline in local model - **fixed** — Respect Traditional Chinese locale in selection assistant ##### What's Changed * feat(chat-branch): persist empty branch nodes by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17667 * perf(chat-tabs): skip redundant re-show work in message tools, artifact pane, and trace page by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18038 * fix(claude-code): reap sdk cli subprocesses on shutdown by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/17969 * fix(tool-loop): raise the default tool-call round cap by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18072 * fix(directory-tree): close the snapshot-to-stream delivery gap by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18009 * feat(providers): built-in web search for DeepSeek/OpenRouter/Dashscope + distinguishable resolved model names by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17983 * fix(v2-migration): allow direct upgrades to v2.0.x by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18093 * feat(cache-cleanup): add selective cache and legacy data cleanup by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/17377 * fix(composer): open image tokens in shared preview by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/18082 * fix(agent-channels): start WeChat QR login by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/18071 * refactor(proxy): make ProxyService own worker routing policy by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18095 * feat: add Claude (Anthropic) conversation import support by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/16164 * feat(cli-config): support detailed models through unified gateway by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18014 * feat(data-migration): allow rerunning v1 migration by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18002 * fix(mcp-settings): preserve form edits and improve tool details by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/17647 * fix(new-api): serve every endpoint from one host with per-route versions by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/18033 * fix(agent-migration): skip symlinks during migration by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18101 * fix(local-model): survive mirror failures and keep embedding inference offline by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/18016 * chore: release v2.0.2 by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/18107 * fix(selection-assistant): respect Traditional Chinese locale by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/18098 **Full Changelog**: https://github.com/CherryHQ/cherry-studio/compare/v2.0.1...v2.0.2 ### v2.0.1 - Date: 2026-08-07 - Version: v2.0.1 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.1 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.1 - **fixed** — Localize workspace tree errors in agent files - **fixed** — Use preset append mode for agent prompts in claude-code - **security** — Address high and critical security advisories - **fixed** — Serve DeepSeek V4 models over the Anthropic endpoint - **fixed** — Recover stale dashboards across windows in openclaw - **fixed** — Prevent reference context leaks in collapsed previews - **fixed** — Normalize no-prefill agent requests in API gateway - **fixed** — Persist provider filter in provider settings - **fixed** — Hide prompt variable hint in agent dialog - **fixed** — Use free Qwen instead of the default assistant model in AI diagnosis - **fixed** — Allow WebP avatar uploads in profile - **fixed** — Isolate nested scroll regions from chat viewport - **fixed** — Hide disabled MCP tools from agents in claude-code - **fixed** — Preserve TeX when copying formulas in selection menu - **fixed** — Add bidirectional controls to horizontal scroll container - **changed** — Reduce renderer loading work in agent session - **fixed** — Show model identity on single replies in chat - **added** — Enable drag reorder for assistant groups - **fixed** — Avoid redundant first-page requests in resource pagination - **fixed** — Normalize internal Agent continuations in API gateway - **added** — Add 'embedding' endpoint type for NEWAPI model configuration - **added** — Keep the create row visible while searching in conversation picker - **added** — Support built-in web search for DeepSeek - **fixed** — Bypass proxy for loopback gateways in agent runtime - **fixed** — Align export and save actions in export menu - **added** — Show settings as a focused tab in tab bar - **fixed** — Strip schema dialect metadata in gemini tools - **fixed** — Add model type scroll controls in provider settings - **fixed** — Tolerate unusable legacy API keys in provider migration - **fixed** — Restore default reasoning effort in composer - **fixed** — Restore normal agent capabilities in cherry assistant - **fixed** — Preserve conversation tab identity in tab navigation - **added** — Draw a Chrome-style divider between adjacent tabs in tab bar - **fixed** — Restore assistant identity for migrated messages in chat - **fixed** — Preserve lazy-loaded data between openings in model selector - **fixed** — Support Claude tools in Responses API in API gateway - **fixed** — Guard query type in WebSearchTool and normalizeWebSearchKeywords, fix translation button contrast - **added** — Read binary office documents through anydoc in knowledge readers - **fixed** — Prevent crash by adding error handling to V2 migration pipeline - **fixed** — Stop repeated saves after failures in resource catalog - **fixed** — Prevent filtered model selection crash in model selector - **fixed** — Migrate tokenflux painting history in painting migrator - **fixed** — Route OpenCode Go models by their served protocol in provider registry ##### What's Changed * fix(agent-files): localize workspace tree errors by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/17869 * fix(claude-code): use preset append mode for agent prompts by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17879 * chore(deps): address high and critical advisories by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/17875 * fix(provider-registry): serve DeepSeek V4 models over the Anthropic endpoint by @eeee0717 in https://github.com/CherryHQ/cherry-studio/pull/17901 * fix(openclaw): recover stale dashboards across windows by @vaayne in https://github.com/CherryHQ/cherry-studio/pull/17864 * fix(chat-messages): prevent reference context leaks in collapsed previews by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17889 * fix(api-gateway): normalize no-prefill agent requests by @kovsu in https://github.com/CherryHQ/cherry-studio/pull/17899 * fix(provider-settings): persist provider filter by @ousugo in https://github.com/CherryHQ/cherry-studio/pull/17866 * fix(agent-dialog): hide prompt variable hint by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17903 * fix(ai-diagnosis): use free Qwen instead of the default assistant model by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/17902 * fix(profile): allow WebP avatar uploads by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17862 * fix(message-scroll): isolate nested scroll regions from chat viewport by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17876 * fix(claude-code): hide disabled MCP tools from agents by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17914 * fix(selection-menu): preserve TeX when copying formulas by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17905 * fix(horizontal-scroll-container): add bidirectional controls by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/17921 * perf(agent-session): reduce renderer loading work by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17878 * fix(chat): show model identity on single replies by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17928 * feat(assistant-groups): enable drag reorder by @zhangjiadi225 in https://github.com/CherryHQ/cherry-studio/pull/17911 * fix(resource-pagination): avoid redundant first-page requests by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17931 * fix(api-gateway): normalize internal Agent continuations by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17929 * chore(deps): upgrade selection-hook to 2.0.3 by @0xfullex in https://github.com/CherryHQ/cherry-studio/pull/17934 * feat: Add 'embedding' endpoint type for NEWAPI model configuration by @LWWZH in https://github.com/CherryHQ/cherry-studio/pull/14988 * feat(conversation-picker): keep the create row visible while searching by @kennyzheng-builds in https://github.com/CherryHQ/cherry-studio/pull/17930 * feat(deepseek): support built-in web search by @DeJeune in https://github.com/CherryHQ/cherry-studio/pull/17913 * fix(agent-runtime): bypass proxy for loopback gateways by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/17906 * fix(export-menu): align export and save actions by @Pleasurecruise in https://github.com/CherryHQ/cherry-studio/pull/17920 * feat(tab-bar): show settings as a focused tab by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/17916 * fix(gemini-tools): strip schema dialect metadata by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/17945 * fix(provider-settings): add model type scroll controls by @AtomsH4 in https://github.com/CherryHQ/cherry-studio/pull/17941 * 🤖 Daily Auto I18N Sync: Aug 06, 2026 by @github-actions[bot] in https://github.com/CherryHQ/cherry-studio/pull/17960 * fix(provider-migration): tolerate unusable legacy API keys by @kangfenmao in https://github.com/CherryHQ/cherry-studio/pull/1 _[Truncated at 4000 characters — full notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.1]_ ### v2.0.0 - Date: 2026-08-05 - Version: v2.0.0 - Original notes: https://github.com/CherryHQ/cherry-studio/releases/tag/v2.0.0 - Permalink: https://whatsnew.fyi/product/cherry-studio/releases/v2.0.0 - **added** — Support for both OpenAI and Anthropic endpoints - **added** — Built-in core Agent runtime, reducing dependency installation and environment setup - **added** — Agents can read and maintain knowledge bases - **added** — Multi-window and split-view layouts for multitasking - **added** — One-click previews of agent results - **added** — One-click model service configuration for Claude Code, Codex, and more in Code Mate - **changed** — Upgraded interactions and visual design for chats, roles, projects, and tasks - **changed** — Code tools upgraded to Code Mate - **changed** — Image generation now dynamically shows available parameters for each model - **changed** — Data foundation upgraded for responsiveness with extensive history - **fixed** — Improved scrolling performance in very long conversations - **fixed** — Large-folder imports no longer block the interface Cherry Studio 2.0.0 - A New AI Workspace, from Chat to Getting Work Done ✨ New Interface, New Experience - Upgraded interactions and visual design make chats, roles, projects, and tasks clearer and easier to organize - Built for multitasking, with multi-window and split-view layouts for switching between work and reviewing progress and deliverables 🤖 More Powerful Agents - Supports both OpenAI and Anthropic endpoints, reuses Claude and Codex subscriptions, and works with more models - The core Agent runtime is built in, greatly reducing dependency installation and environment setup - Complete research, documents, presentations, data analysis, and coding faster, with one-click previews of results 🛠️ More Tool Upgrades - Agents can read and maintain knowledge bases, helping your knowledge grow continuously like a Wiki - Code tools are upgraded to Code Mate, with one-click model service configuration for Claude Code, Codex, and more - Image generation dynamically shows the parameters available for each model, making results easier to control ⚡ Performance - A fully upgraded data foundation stays responsive even with extensive history - Smoother scrolling in very long conversations, while large-folder imports no longer block the interface