5 versions of TextLens are listed here, from July 24, 2026 to the current 0.4.5. Each one links to the publisher’s own installer files.
| Version | Date | Size | Files |
|---|---|---|---|
| 0.4.5Latest | November 9, 2026 | — | 64-bit EXE (NSIS) Release notes- Improved macOS selection compatibility with screenshot overlays. Clipboard fallback no longer injects copy shortcuts when an overlay covers the selection gesture. Moved host accessibility queries off the main thread to keep result-window updates responsive.
- Improved streaming results. The first text chunk appears immediately, later updates are batched per frame, and hidden-window timers ensure the stream continues to flush. Refined scrolling, waiting, copying, and Markdown presentation.
- Added model-aware thinking controls for each tool. Detection uses model, protocol, endpoint, and capability metadata for OpenAI Chat/Responses, Anthropic, Gemini, OpenRouter, and common Qwen, GLM, Kimi, and DeepSeek models. Added the max level; when a model cannot disable thinking, selecting off uses its lowest supported level.
- Moved thinking settings to each tool’s model selection, with controls to enable or disable thinking, select a level, and rerun automatic detection. Different tools can use different levels for the same model.
- Improved the default translation and explanation prompts. Translation better preserves meaning, tone, protected content, and document structure. Explanations can use necessary field terminology while remaining accessible. Only untouched built-in prompts are upgraded.
- Improved language detection with shared TypeScript and Rust rules. Code, URLs, commands, and logs are excluded from language detection without changing the text sent to the model. Fixed inline-code lines being mistaken for fenced code.
- Improved PDF text handling by merging semantically continuous wrapped lines into paragraphs while preserving paragraph, list, table, and code boundaries.
发布版本 0.4.5
- 提升 macOS 划词与截图遮罩的兼容性。截图浮层遮住手势位置时,不再向宿主注入复制快捷键;将宿主辅助功能查询移出主线程,保持结果窗口响应流畅。
- 改进流式结果显示:首段文本立即出现,后续增量按帧合并,并通过隐藏窗口定时器确保内容持续刷新;优化滚动、等待、复制和 Markdown 呈现。
- 为各工具加入模型感知的思考控制。结合模型、协议、端点和能力元数据识别 OpenAI Chat/Responses、Anthropic、Gemini、OpenRouter 及常见 Qwen、GLM、Kimi、DeepSeek 模型。增加 max 档位;模型不支持关闭思考时,选择关闭会使用其最低支持档位。
- 将思考设置移至各工具的模型选择处,支持启用或关闭思考、选择档位和重新自动识别。同一模型可按工具分别设置不同档位。
- 优化默认翻译与解释提示词。翻译更注重保留语义、语气、受保护内容和文档结构;解释可使用必要的学科术语,同时保持易懂。仅更新未被用户编辑的内置提示词。
- 提高语言识别准确度,TypeScript 与 Rust 共用识别规则。识别时过滤代码、URL、命令和日志,但不改动发送给模型的原文;修复行内代码被误判为代码围栏的问题。
- 改进 PDF 折行文本处理,自动合并语义连续的文本行并恢复段落,同时保留段落、列表、表格和代码边界。 |
| 0.4.3 | November 6, 2026 | — | 64-bit EXE (NSIS) Release notes- Fixed bugs and logic errors found during a source review, covering selection handling, action execution, settings, and result windows across macOS and Windows.
- Improved toolbar and result-window opening speed by running initialization tasks in parallel. Added a shorter, smoother 90 ms fade-in transition.
- Improved streaming result playback and Markdown rendering. The display adapts to incoming content and shows the complete response when generation finishes.
- Improved Ask AI retry behavior. Retrying a failed response preserves the question and conversation history.
- Improved settings-save reliability so edits made during saving remain in the draft. Refined dialog focus handling and navigation to fields with validation errors.
- Prevented actions bound to disabled providers from running. On macOS, selection highlights are cleared after copying or searching succeeds, or after the result window appears successfully.
发布版本 0.4.3
- 修复源码审查中发现的潜在 bug 与逻辑错误,覆盖 macOS 和 Windows 的划词处理、动作执行、设置保存及结果窗口等功能。
- 优化工具栏和结果框的弹出速度,并行处理窗口初始化任务;采用更短、更流畅的 90 毫秒淡入过渡。
- 优化流式结果播放与 Markdown 渲染,显示节奏随内容到达情况自适应调整,并在生成结束后补齐完整响应。
- 改进问 AI 的失败重试行为,重试时保留原问题及对话历史。
- 提升设置保存可靠性,保留保存期间继续编辑的新草稿;改进弹窗焦点处理与校验错误字段定位。
- 阻止已绑定到停用服务商的动作继续执行;macOS 在复制或搜索成功、或结果窗口成功显示后,再延迟清除宿主应用中的选区高亮。 |
| 0.4.2 | November 1, 2026 | — | 64-bit EXE (NSIS) Release notes- Improved selection capture and stability across macOS and Windows, including app-specific handling, ordered event processing, and more reliable recovery when the toolbar or settings renderer encounters an error.
- Prevented stale selections from reopening the toolbar. After an action, TextLens clears the highlight in the app that owns the selection; if the clipboard cannot be fully restored, temporary copying is skipped to protect images and other clipboard content.
- Rebuilt Settings into seven sections, added field-level error feedback and filter-tag editing, and made saving run in the background with backups. Testing a connection or fetching models now prompts you to save pending changes first.
- Added OpenAI Responses, Anthropic Messages, and Google Gemini protocols alongside OpenAI Chat Completions, with support for streaming and model discovery.
- Expanded translation to seven languages: Chinese, English, Japanese, Korean, Russian, German, and French. Added configurable default and target languages, automatic direction switching, and a control to swap translation direction.
- Added per-model thinking controls, including selectable thinking levels and an option to disable thinking when the model supports it. Models that require thinking use their minimum budget when thinking is turned off.
- Added five appearance themes and improved keyboard navigation and focus indicators across menus and controls. Ask AI results now copy the complete conversation transcript.
- Improved streaming responsiveness and input-limit handling. Diagnostic logs show selection length and a hash by default; text previews require explicit opt-in.
发布版本 0.4.2
- 提升 macOS 与 Windows 的划词捕获稳定性,优化应用适配与事件处理顺序,并改进工具栏或设置界面异常后的恢复。
- 改进旧选区判定,避免工具栏重复弹出;执行动作后清除选区所属应用中的高亮。剪贴板无法完整恢复时跳过临时复制,保护截图等内容。
- 设置页重构为七个分区,加入字段级错误提示和过滤标签编辑;设置保存改为后台处理并保留备份。测试连接或获取模型前,会提示先保存未提交的更改。
- 在 OpenAI Chat Completions 之外,新增 OpenAI Responses、Anthropic Messages 和 Google Gemini 协议支持,并支持流式输出与模型列表获取。
- 翻译扩展至中文、英语、日语、韩语、俄语、德语和法语;支持设置默认语言与目标语言、自动切换翻译方向,以及手动交换方向。
- 新增按模型配置思考强度,可选择思考档位;模型支持时可关闭思考。无法完全关闭思考的模型会使用最低预算。
- 新增五套外观主题,改进菜单与控件的键盘导航和焦点提示;复制问 AI 结果时会包含完整对话记录。
- 提升流式输出响应并完善输入限制。诊断日志默认只记录选区长度和哈希;文本预览需显式开启。 |
| 0.4.0 | August 10, 2026 | — | 64-bit EXE (NSIS) Release notes- Improved selection capture compatibility and stability across macOS and Windows, including PDF, Office, browsers, and custom-drawn text areas
- Faster and more reliable selection toolbar with staged presentation, hover feedback, and long-running recovery
- More stable streaming Markdown rendering with consistent spacing for paragraphs, lists, code blocks, tables, and math
- Improved global shortcut registration, rollback, modifier-key capture, and multi-turn follow-up support |
| 0.3.56 | July 24, 2026 | — | 64-bit EXE (NSIS) |