TeXada

English

TeXada

TeXada is an on-device, agent-driven structured math editor for converting natural language, partial LaTeX, and screenshots into usable formula blocks. It deliberately has only two model roles: MiniCPM5-1B handles planning and text generation, while MiniCPM-V 4.6 handles image understanding and formula OCR. Independently testable TeX tools own parsing, validation, deterministic repair, semantic diffing, rendering and export.

MiniCPM5 Planner → TeX Tool → Observation → MiniCPM5 Planner
                       |
                       v
                 Semantic Unit

MiniCPM5 does not repair invalid formulas itself. repair_tex is a deterministic local syntax tool, not a third model. See Architecture, Design Evolution, and Local E2E.

The original SymbolEngine and Operator-Drift Guard remain Level 0: they pin critical operators with near-zero overhead. A pinned KaTeX AST bridge and role-aware weighted Semantic Diff form Level 1 when an explicit before/after formula exists. Raw natural language is never treated as a reference AST.

Highlights

Capability	Description
Natural language to LaTeX	Describe a formula and get copyable LaTeX or Markdown
Formula OCR	MiniCPM-V proposes a formula; MiniCPM5 reviews it through TeX tools
Completion	Rules or MiniCPM5 propose a completion; the shared Agent validates it
Validation and repair	Check, fix, render and highlight LaTeX
Presets, history, and run logs	Keep reusable presets, reopen results, and inspect each Agent/OCR/completion success or failure with model, latency, tools, and trace
Desktop insertion	Click a formula block to type it at the system cursor
UI controls	Switch language, zoom from 80% to 140%, and drag the floating window
Release packages	macOS DMGs and Windows x64 NSIS installers from GitHub Actions

Agent Technical Highlights

Layer	Technical highlight
MiniCPM5-native planning	Accepts MiniCPM5 XML tool calls and normalized OpenAI `tool_calls` without inventing a separate Agent protocol
Six specialist tools	`parse_tex`, `compile_tex`, deterministic `repair_tex`, `semantic_diff`, `render_math`, and `export` have narrow schemas and independent tests
Semantic state	Pinned KaTeX 0.17.0 runs in reusable in-process V8 and is normalized into fractions, roots, integrals, scripts, matrix rows/cells, and other Semantic Units
Two-level guard	Level 0 preserves critical operators and integral rank; Level 1 compares explicit before/after formulas with role-aware weighted structural Diff
Bounded execution	Step limits, repeated-call fingerprints, consecutive-error cutoffs, operator-drift retry limits, request timeouts, and final compile/render guards prevent small-model loops
Deterministic acceleration	High-confidence NL and common completion prefixes take a zero-token path while still producing real compile/render observations
Unified inputs	NL, MiniCPM-V OCR candidates, and completion candidates share one runtime, trace format, stop reasons, and result contract
Observable local runs	A request ledger records run ID, models, latency, tokens, tools, stop reason, formula validity, errors, and expandable Agent trace

TeXada deliberately does not embed a TeX2TeX repair model. Formula repair is a deterministic specialist tool, while experimental repair models can evolve in a separate research repository without coupling product runtime or release size.

The Ollama port is configurable. The default is http://localhost:11434, but Settings, ~/.texada/config.json, and TEXADA_OLLAMA_HOST can point TeXada at any host or port.

Download

Version 0.3.2 is released from the main branch.

Platform	Package
macOS Apple Silicon	`TeXada_0.3.2_aarch64.dmg`
macOS Intel	`TeXada_0.3.2_x64.dmg`
Windows x64	`TeXada_0.3.2_x64-setup.exe`

Release page: github.com/CacinieP/TeXada-the-Math-Agent/releases

From A Fresh Machine To First Formula

TeXada release packages are built for end users. You do not need Python, Node.js, Rust, a source checkout, or a separate FastAPI server.

Download the package for your machine from the release page.
- Apple Silicon Macs use the aarch64.dmg.
- Intel Macs use the x64.dmg.
- Windows PCs use the x64-setup.exe.
Install and open TeXada.
- macOS: open the DMG and drag TeXada into Applications.
- Windows: run the setup EXE and start TeXada from the Start menu.
Pick one model path:
- Local path: install Ollama and pull the default text model. Pull the vision model too if you want OCR.
- Cloud path: skip Ollama, open Settings, choose OpenAI-compatible, enter the provider base URL, text model, vision model and API key, then save.
Generate the first formula.
- Open the NL tab.
- Type integral of x squared from 0 to 1.
- Press Enter to run the conversion.
- Use the copy button, or place the cursor in another app and click the formula block to type it there.
Try OCR after the vision model is ready.
- Paste or drop a screenshot into the OCR tab.
- If the title bar says Text ready · OCR missing, text conversion still works; pull or configure the vision model before OCR.
Reuse previous work from History.
- Open the History tab to search saved natural-language, completion, OCR, and LaTeX results.
- Use the type filter to focus on Natural, Complete, or OCR records.
- Click Reuse input to send a saved natural-language or completion prompt back to the matching tab for editing and rerun.
- Switch to Run logs to filter request-level records and expand execution details.

Quick Start With Ollama

Install Ollama from ollama.com/download.
- macOS: use the official download app.
- Windows: use the official Windows installer, then launch Ollama once from the Start menu.
Pull the default local models. The text model is required for local text conversion; the vision model is required only for local OCR:

ollama pull hf.co/openbmb/MiniCPM5-1B-GGUF:Q4_K_M
ollama pull openbmb/minicpm-v4.6:latest

Open TeXada from the downloaded .dmg or .exe. The packaged app includes the TeXada FastAPI backend and starts it automatically; no separate Python install or manual API server is needed.
Check the status in the title bar.
- Ready: text conversion and OCR are available.
- Text ready · OCR missing: text conversion works; pull the vision model shown in the status tooltip.
- Model missing: pull the text model shown in the status tooltip.
- Disconnected: start Ollama or check the configured port.

Connection Map And Ports

TeXada uses two local HTTP layers. They should normally use different ports:

Desktop UI -> TeXada FastAPI API -> Ollama or cloud model endpoint

Layer	Default	What it is for
TeXada FastAPI API	`http://127.0.0.1:18732`	Bundled with the installer and auto-started by the desktop app
Ollama model endpoint	`http://localhost:11434`	The FastAPI backend calls local model APIs and adds `/v1` internally
Cloud model endpoint	Provider-specific `/v1` base URL	Used only when Backend is `OpenAI-compatible`

Do not set the FastAPI address and Ollama address to the same port unless you are deliberately running a custom proxy. If Settings shows a network/API error, restart TeXada and check whether 18732 is already occupied. If FastAPI is reachable but status says Disconnected or Model missing, check Ollama and the model names.

Role	Default	Notes
Text	`hf.co/openbmb/MiniCPM5-1B-GGUF:Q4_K_M`	Natural language conversion and completion
Vision	`openbmb/minicpm-v4.6:latest`	OCR from screenshots and images

These are the only two supported model roles. OpenAI-compatible configuration is a transport option for serving the same MiniCPM models through vLLM, SGLang, or another compatible endpoint; it does not add a third product model.

Ollama does not have to run on port 11434. In Settings → Backend → Ollama address, use any reachable endpoint:

http://localhost:11435
http://192.168.1.20:11434

Do not add /v1; TeXada adds the OpenAI-compatible suffix internally. For local loopback endpoints, TeXada can try to start ollama serve if the Ollama CLI is installed. Remote Ollama endpoints must already be running and reachable.

Cloud Mode

OpenAI-compatible models can be configured from Settings. For cloud providers, enter the complete Chat Completions base URL expected by that provider; TeXada does not append /v1 to cloud endpoints. Example:

{
  "backend": "openai_compatible",
  "openai_base_url": "https://your-provider.example/v1",
  "openai_model_name": "your-text-model",
  "openai_vision_model_name": "your-vision-model",
  "openai_api_key": "your-api-key"
}

StepFun Step Plan example:

{
  "backend": "openai_compatible",
  "openai_base_url": "https://api.stepfun.com/step_plan/v1",
  "openai_model_name": "step-3.7-flash",
  "openai_vision_model_name": "step-3.7-flash",
  "openai_api_key": "your-api-key"
}

The StepFun Step Plan OpenAI-compatible Chat Completions base URL is https://api.stepfun.com/step_plan/v1, and step-3.7-flash is the recommended validation model in StepFun's public documentation.

Configuration

Persistent config lives at ~/.texada/config.json.

{
  "backend": "ollama",
  "ollama_host": "http://localhost:11434",
  "model_name": "hf.co/openbmb/MiniCPM5-1B-GGUF:Q4_K_M",
  "vision_model_name": "openbmb/minicpm-v4.6:latest",
  "api_host": "127.0.0.1",
  "api_port": 18732,
  "ui_language": "zh",
  "ui_zoom": 1.0,
  "max_tokens": 2048,
  "run_log_max_days": 0,
  "run_log_max_items": 0
}

Variable	Purpose
`TEXADA_BACKEND`	`ollama` or `openai_compatible`
`TEXADA_OLLAMA_HOST`	Local Ollama base URL, including custom ports
`TEXADA_MODEL_NAME`, `TEXADA_VISION_MODEL_NAME`	Local text and vision model names
`TEXADA_OPENAI_BASE_URL`	Full OpenAI-compatible cloud base URL
`TEXADA_OPENAI_MODEL_NAME`, `TEXADA_OPENAI_VISION_MODEL_NAME`	Cloud text and vision model names
`TEXADA_OPENAI_API_KEY`	Cloud provider API key
`TEXADA_API_HOST`, `TEXADA_API_PORT`	FastAPI bind address
`TEXADA_API_BASE`	Explicit desktop shell API base
`TEXADA_DISABLE_BUNDLED_BACKEND`	Set to `1` only when you want to manage the FastAPI backend yourself
`TEXADA_API_TIMEOUT_SECS`	Desktop API request timeout
`TEXADA_INFERENCE_TIMEOUT_SECONDS`, `TEXADA_API_REQUEST_TIMEOUT_SECONDS`	Backend model and HTTP timeouts
`TEXADA_AGENT_MAX_STEPS`	Maximum MiniCPM5 planner/tool turns
`TEXADA_UI_LANGUAGE`, `TEXADA_UI_ZOOM`	UI language (`zh` or `en`) and zoom (`0.8` to `1.4`)

Data Backup

TeXada can export and import local data from Settings → Data.

Exports are JSON files and can include conversion history, request-level run logs, user-defined presets, and non-sensitive settings. API keys and OCR image bytes are never exported. History, logs, and presets can also be exported or imported independently. Imports merge by default and skip duplicates.

See docs/data-backup.md for the JSON format.

Privacy And Limits

Topic	Behavior
Local Ollama mode	Model requests stay on the configured Ollama endpoint
Cloud mode	Text and images are sent to the configured provider endpoint
API keys	Saved in `~/.texada/config.json`; do not paste keys into public issues
OCR uploads	Accepts PNG, JPEG, WebP, BMP and TIFF up to 5 MB
Math correctness	Always review generated formulas before using them in final work

Troubleshooting

Symptom	Check
Settings shows API/network error	Restart TeXada; if it persists, check whether another process already uses `127.0.0.1:18732`
Status is `Disconnected`	Ollama is not running or `ollama_host` points to the wrong host/port
Status is `Model missing`	Pull the text model shown in the status tooltip
Status is `Text ready · OCR missing`	Pull the vision model shown in the status tooltip
Cloud mode returns 401/403	API key, provider base URL, and model name
Clicking a formula copies instead of inserting	Browser mode falls back to copy; desktop insertion also needs OS paste automation permission

Experimental CAS Boundary

Version 0.3.2 includes an optional developer-facing CAS scaffold and a reproducible capability matrix. It is not connected to the Agent, API, or desktop UI, and the six public TeX tools remain unchanged. TeXada therefore does not claim to prove arbitrary formula correctness in this release.

The production-direction adapter starts from TeXada Semantic Units and accepts only a declared scalar subset. Raw ANTLR/Lark LaTeX parsing exists only in the evaluation probes. Unsupported notation is rejected, .equals() == False never becomes a contradiction by itself, and every positive result records its assumptions and evidence path.

Source contributors can run the pinned gate with:

uv run --extra dev --extra cas-eval pytest \
  tests/test_cas_translator.py \
  tests/test_cas_checker.py \
  tests/test_cas_capabilities.py

See the SymPy capability matrix for the supported boundary, known parser drift, seed policy, and acceptance red lines.

Release CI

GitHub Actions builds release installers manually from main and automatically from version tags.

Workflow	Checks
`Audit`	Ruff, pytest, pip-audit, npm audit, JS syntax check, Tauri cargo check on macOS and Windows
`Desktop Release`	Pre-release audit, PyInstaller FastAPI sidecar, signed and notarized macOS arm64/Intel DMG, Windows x64 NSIS installer, official release publishing

Shortcuts

Action	Shortcut
Show or hide popup	macOS `Option+Command+T`, Windows `Ctrl+Alt+T`
Toggle render mode	macOS `Command+K`, Windows `Ctrl+K`
Zoom in	`Command/Ctrl + +`
Zoom out	`Command/Ctrl + -`
Reset zoom	`Command/Ctrl + 0`
Move window	drag the header or empty panel space

Hardware And Measurements

The numbers below are real local measurements from 2026-07-07, not synthetic benchmark data. Latency depends on model size, quantization, machine load and whether the model is warm.

Measured environment: Mac Neo, macOS 26.5.1, arm64, Apple A18 Pro, 8GB RAM, local Ollama models.

Scenario	Recommended hardware
Text conversion and completion	Apple Silicon, Intel i5/Ryzen 5 or better, 8GB RAM
Regular screenshot OCR	Apple Silicon M2/M3 class or better, 16GB RAM
Heavy OCR or larger models	16GB+ RAM, discrete GPU, or a cloud vision model

Operation	Model	Observed latency
NL to LaTeX, cold	MiniCPM5-1B	179204.3ms
NL to LaTeX, warm	MiniCPM5-1B	29612.2ms
LaTeX completion	MiniCPM5-1B plus rules	1808.5ms
OCR sample	MiniCPM-V 4.6	39419.0ms

Quality Gate

The current main branch has no known blocking issue after the latest audit pass. That means tests and static checks pass; it is not a mathematical promise that no software bug can exist. Maintainer validation details live in Contributing, Source audit, and GitHub Actions.

Documentation

Area	Links
Technical docs	Docs index, Architecture, Technical report, Source audit, File inventory
Community	Contributing, Security, Support, Code of Conduct, Changelog

License

TeXada-the-Math-Agent is released under GPL-3.0-or-later. See LICENSE.

中文

TeXada

TeXada 是一个基于端侧 Agent 的结构化数学编辑器，用于把自然语言、不完整 LaTeX 和截图转为可用公式块。产品明确只有两个模型角色：MiniCPM5-1B 负责规划、工具选择和文本生成，MiniCPM-V 4.6 负责图片理解与公式 OCR；解析、校验、确定性修复、语义 Diff、渲染与导出由可独立测试的 TeX 工具负责。

MiniCPM5 Planner → TeX Tool → Observation → MiniCPM5 Planner
                       |
                       v
                 Semantic Unit

MiniCPM5 不直接修复非法公式；repair_tex 是确定性本地语法工具，不是第三个模型。详见架构文档与设计思路与版本迭代、本地 E2E 指南。

原有 SymbolEngine 与 Operator-Drift Guard 被保留为 Level 0，以近乎零开销锁定关键算符；固定版本的 KaTeX AST 桥与角色感知的加权 Semantic Diff 构成 Level 1，但只在明确存在 before/after 公式时启用，不会把自然语言伪装成参考 AST。

亮点

能力	说明
自然语言转 LaTeX	输入公式描述，生成可复制的 LaTeX 或 Markdown
公式 OCR	MiniCPM-V 产出候选，MiniCPM5 再通过 TeX 工具审查
公式补全	规则或 MiniCPM5 产出候选，再由统一 Agent 校验
校验与修复	检查、修复、渲染并高亮 LaTeX
预设、历史与运行日志	保存预设、重新打开结果，并检查每次 Agent/OCR/补全成功或失败的模型、耗时、工具与轨迹
桌面键入	点击公式块即可在系统当前光标处键入公式
界面控制	设置页切换中英文、80% 到 140% 缩放、拖动浮窗
安装包发布	GitHub Actions 构建 macOS DMG 和 Windows x64 NSIS 安装包

Agent 技术亮点

层级	技术亮点
MiniCPM5 原生规划	同时接收 MiniCPM5 XML tool calling 与标准化 OpenAI `tool_calls`，不另造一套 Agent 协议
六个专家工具	`parse_tex`、`compile_tex`、确定性 `repair_tex`、`semantic_diff`、`render_math`、`export` 均有窄 schema 和独立测试
结构化状态	固定 KaTeX 0.17.0 在可复用进程内 V8 中运行，并归一化为分式、根式、积分、上下标、矩阵行列等 Semantic Unit
两级守卫	Level 0 保留关键算符和积分阶数；Level 1 对明确的 before/after 公式执行角色感知加权结构 Diff
有界执行	步数上限、重复调用指纹、连续错误截断、算符漂移重试上限、请求超时与最终 compile/render guard 防止小模型循环
确定性加速	高置信度 NL 和常见补全前缀走零 token 路径，但仍生成真实 compile/render Observation
三入口统一	NL、MiniCPM-V OCR 候选、补全候选共用 Runtime、trace、stop reason 与结果合约
本地可观察性	请求账本记录 run ID、模型、耗时、token、工具、停止原因、公式有效性、错误和可展开 Agent trace

TeXada 不内置 TeX2TeX 修复模型。公式修复由确定性专家工具负责，实验性修复模型可在独立研究仓库演进，不与产品 Runtime 和安装包体积耦合。

Ollama 端口不是写死的。默认地址是 http://localhost:11434，但可以在设置页、~/.texada/config.json 或 TEXADA_OLLAMA_HOST 中改为任意主机和端口。

下载

版本 0.3.2 从 main 分支发布。

平台	安装包
macOS Apple Silicon	`TeXada_0.3.2_aarch64.dmg`
macOS Intel	`TeXada_0.3.2_x64.dmg`
Windows x64	`TeXada_0.3.2_x64-setup.exe`

Release 页面：github.com/CacinieP/TeXada-the-Math-Agent/releases

从裸机到第一次使用

TeXada 的 release 安装包面向普通用户。你不需要安装 Python、Node.js、Rust，也不需要拉源码或手动启动 FastAPI 服务。

从 Release 页面下载适合自己机器的安装包。
- Apple Silicon Mac 选择 aarch64.dmg。
- Intel Mac 选择 x64.dmg。
- Windows 电脑选择 x64-setup.exe。
安装并打开 TeXada。
- macOS：打开 DMG，把 TeXada 拖到 Applications。
- Windows：运行安装器，然后从开始菜单启动 TeXada。
选择一种模型路线。
- 本地路线：安装 Ollama，并拉取默认文本模型；如果要用 OCR，再拉取视觉模型。
- 云侧路线：可以跳过 Ollama；进入设置页，选择 OpenAI-compatible，填写 provider base URL、文本模型、视觉模型和 API key，然后保存。
生成第一个公式。
- 打开 NL 页签。
- 输入 0 到 1 上 x 平方的积分。
- 按 Enter 执行转换。
- 可以点击复制按钮，也可以先把光标放到其他应用里，再点击公式块把公式键入到光标处。
视觉模型准备好后再试 OCR。
- 在 OCR 页签粘贴或拖入公式截图。
- 如果标题栏显示 文本可用 · OCR 缺模型，文本转换仍可用；OCR 需要先拉取或配置视觉模型。
从历史页复用旧记录。
- 打开 历史 页签，搜索保存过的自然语言、补全、OCR 和 LaTeX 结果。
- 使用类型筛选专注查看自然语言、补全或 OCR 记录。
- 点击 复用输入 可以把历史自然语言或补全片段送回对应页签，继续编辑或重新生成。
- 切换到 运行日志，可以筛选请求并展开查看执行详情。

Ollama 快速启动

从 ollama.com/download 安装 Ollama。
- macOS：使用官方下载版应用。
- Windows：使用官方 Windows 安装器，安装后先从开始菜单启动一次 Ollama。
拉取默认本地模型。文本模型用于本地文本转换；视觉模型只在本地 OCR 时必需：

ollama pull hf.co/openbmb/MiniCPM5-1B-GGUF:Q4_K_M
ollama pull openbmb/minicpm-v4.6:latest

打开下载好的 TeXada .dmg 或 .exe 安装包。安装包内置 TeXada FastAPI 后端，并会在应用启动时自动拉起；不需要单独安装 Python 或手动启动 API 服务。
看标题栏状态。
- Ready：文本转换和 OCR 都可用。
- 文本可用 · OCR 缺模型：文本可用，按状态 tooltip 里的命令拉取视觉模型。
- 模型缺失：按状态 tooltip 里的命令拉取文本模型。
- 未连接：启动 Ollama，或检查配置的端口。

连接层级与端口

TeXada 使用两层本地 HTTP 服务。它们通常应该是不同端口：

桌面 UI -> TeXada FastAPI API -> Ollama 或云侧模型 endpoint

层级	默认值	用途
TeXada FastAPI API	`http://127.0.0.1:18732`	随安装包内置，由桌面应用自动启动
Ollama 模型 endpoint	`http://localhost:11434`	FastAPI 后端调用本地模型接口，并在内部自动拼接 `/v1`
云侧模型 endpoint	服务商提供的 `/v1` base URL	仅在后端选择 `OpenAI-compatible` 时使用

除非你主动配置了自定义代理，不要把 FastAPI 地址和 Ollama 地址设成同一个端口。如果设置页显示 API/网络错误，先重启 TeXada，并检查 18732 是否已被其他进程占用；如果 FastAPI 可达但状态是 未连接 或 模型缺失，再检查 Ollama 和模型名。

角色	默认模型	说明
文本	`hf.co/openbmb/MiniCPM5-1B-GGUF:Q4_K_M`	自然语言转换和补全
视觉	`openbmb/minicpm-v4.6:latest`	从截图和图片识别公式

产品只支持这两个模型角色。OpenAI-compatible 配置只是让同一组 MiniCPM 模型可以通过 vLLM、SGLang 或其他兼容 endpoint 部署，不会引入第三个产品模型。

Ollama 不必固定在 11434。在设置页 → 后端连接 → Ollama 地址里，可以填任意可访问地址：

http://localhost:11435
http://192.168.1.20:11434

不用加 /v1，TeXada 会在内部自动拼接 OpenAI-compatible 后缀。对于本机 loopback 地址，如果已安装 Ollama CLI，TeXada 可以尝试启动 ollama serve；远端 Ollama 地址需要你先在远端启动并保证网络可达。

云侧模式

OpenAI API 兼容模型可以在设置页配置。云侧 provider 需要填写该服务商要求的完整 Chat Completions base URL；TeXada 不会为云侧 endpoint 自动追加 /v1。示例：

{
  "backend": "openai_compatible",
  "openai_base_url": "https://your-provider.example/v1",
  "openai_model_name": "your-text-model",
  "openai_vision_model_name": "your-vision-model",
  "openai_api_key": "your-api-key"
}

StepFun Step Plan 的 OpenAI-compatible Chat Completions base URL 是 https://api.stepfun.com/step_plan/v1，step-3.7-flash 是 StepFun 公开文档里的推荐验证模型。

StepFun Step Plan 示例：

{
  "backend": "openai_compatible",
  "openai_base_url": "https://api.stepfun.com/step_plan/v1",
  "openai_model_name": "step-3.7-flash",
  "openai_vision_model_name": "step-3.7-flash",
  "openai_api_key": "your-api-key"
}

配置文件

持久配置文件位于 ~/.texada/config.json。

{
  "backend": "ollama",
  "ollama_host": "http://localhost:11434",
  "model_name": "hf.co/openbmb/MiniCPM5-1B-GGUF:Q4_K_M",
  "vision_model_name": "openbmb/minicpm-v4.6:latest",
  "api_host": "127.0.0.1",
  "api_port": 18732,
  "ui_language": "zh",
  "ui_zoom": 1.0,
  "max_tokens": 2048
}

环境变量	用途
`TEXADA_BACKEND`	`ollama` 或 `openai_compatible`
`TEXADA_OLLAMA_HOST`	本地 Ollama 地址，支持自定义端口
`TEXADA_MODEL_NAME`, `TEXADA_VISION_MODEL_NAME`	本地文本模型和视觉模型名称
`TEXADA_OPENAI_BASE_URL`	完整 OpenAI-compatible 云侧 base URL
`TEXADA_OPENAI_MODEL_NAME`, `TEXADA_OPENAI_VISION_MODEL_NAME`	云侧文本模型和视觉模型名称
`TEXADA_OPENAI_API_KEY`	云侧 provider API key
`TEXADA_API_HOST`, `TEXADA_API_PORT`	FastAPI 监听地址
`TEXADA_API_BASE`	桌面壳使用的完整 API 地址
`TEXADA_DISABLE_BUNDLED_BACKEND`	仅当你想自行管理 FastAPI 后端时设为 `1`
`TEXADA_API_TIMEOUT_SECS`	桌面端 API 请求超时时间
`TEXADA_INFERENCE_TIMEOUT_SECONDS`, `TEXADA_API_REQUEST_TIMEOUT_SECONDS`	后端模型推理和 HTTP 请求超时
`TEXADA_AGENT_MAX_STEPS`	MiniCPM5 Planner/Tool 最大执行轮数
`TEXADA_RUN_LOG_MAX_DAYS`, `TEXADA_RUN_LOG_MAX_ITEMS`	运行日志保留上限；默认 `0` 表示不限，完整保留每次运行
`TEXADA_UI_LANGUAGE`, `TEXADA_UI_ZOOM`	界面语言（`zh` 或 `en`）和缩放（`0.8` 到 `1.4`）

数据备份

TeXada 可在设置 → 数据中导出和导入本地数据。

导出的 JSON 文件可包含转换历史、请求级运行日志、用户自定义预设和非敏感设置。 API Key 与 OCR 图片原始字节不会被导出。历史、日志与预设也可以单独导入导出；导入默认合并并跳过重复项。运行日志默认不限量保留；列表分页加载轻量摘要，展开某条记录时才读取完整 Agent 轨迹。

JSON 格式见 docs/data-backup.md。

隐私与限制

项目	行为
本地 Ollama 模式	模型请求只发送到配置的 Ollama endpoint
云侧模式	文本和图片会发送到你配置的 provider endpoint
API key	保存在 `~/.texada/config.json`；不要粘贴到公开 issue
OCR 上传	支持 PNG、JPEG、WebP、BMP、TIFF，最大 5 MB
数学正确性	生成公式用于正式内容前仍需人工核对

故障排查

现象	检查项
设置页显示 API/网络错误	重启 TeXada；如果仍存在，检查是否有其他进程占用 `127.0.0.1:18732`
状态是 `未连接`	Ollama 没启动，或 `ollama_host` 主机/端口不对
状态是 `模型缺失`	按状态 tooltip 拉取文本模型
状态是 `文本可用 · OCR 缺模型`	按状态 tooltip 拉取视觉模型
云侧模式返回 401/403	API key、provider base URL 和模型名
点击公式只复制、不键入	浏览器模式会退化为复制；桌面键入还需要系统粘贴自动化权限

实验性 CAS 能力边界

0.3.2 包含面向开发者的可选 CAS 骨架与可复现能力矩阵，但尚未接入 Agent、API 或桌面 UI；公开的六个 TeX 工具保持不变。因此，本版本不宣称能够证明任意公式的数学正确性。

生产方向的 adapter 只接收 TeXada Semantic Unit，并仅转换明确声明的标量子集。 ANTLR/Lark 原始 LaTeX parser 只用于评测探针。未支持的记号会明确拒绝； .equals() == False 不会独自升级为判错；每个正向结论都会记录 assumptions 和证据路径。

源码贡献者可运行固定环境的门禁：

uv run --extra dev --extra cas-eval pytest \
  tests/test_cas_translator.py \
  tests/test_cas_checker.py \
  tests/test_cas_capabilities.py

支持边界、已知 parser 漂移、seed 策略与验收红线见 SymPy 能力矩阵。

Release CI

GitHub Actions 可以手动从 main 构建安装包，也会在版本 tag 上自动构建。

Workflow	检查内容
`Audit`	Ruff、pytest、pip-audit、npm audit、JS 语法检查、macOS/Windows Tauri cargo check
`Desktop Release`	预发布审计、PyInstaller FastAPI sidecar、已签名并公证的 macOS arm64/Intel DMG、Windows x64 NSIS 安装器、正式 release 发布

快捷键

操作	快捷键
显示或隐藏浮窗	macOS `Option+Command+T`，Windows `Ctrl+Alt+T`
切换渲染模式	macOS `Command+K`，Windows `Ctrl+K`
放大	`Command/Ctrl + +`
缩小	`Command/Ctrl + -`
重置缩放	`Command/Ctrl + 0`
移动窗口	拖动标题栏或面板空白区域

硬件与实测

下面是 2026-07-07 的本地实测，不是合成跑分。响应时间会受模型大小、量化方式、机器负载和模型是否已预热影响。

实测环境：Mac Neo，macOS 26.5.1，arm64，Apple A18 Pro，8GB RAM，本地 Ollama 模型。

场景	推荐硬件
文本转换和补全	Apple Silicon、Intel i5/Ryzen 5 或更高，8GB RAM
常用截图 OCR	Apple Silicon M2/M3 级别或更高，16GB RAM
高频 OCR 或更大模型	16GB+ RAM、独立 GPU，或云侧视觉模型

操作	模型	实测响应
NL 转 LaTeX，冷启动	MiniCPM5-1B	179204.3ms
NL 转 LaTeX，预热后	MiniCPM5-1B	29612.2ms
LaTeX 补全	MiniCPM5-1B 加规则	1808.5ms
OCR 示例	MiniCPM-V 4.6	39419.0ms

质量门禁

当前 main 分支在最新审计后没有已知阻塞问题。这里的含义是测试和静态检查通过，不是声称软件绝对不可能有 bug。维护者验证细节见贡献指南、源码审计和 GitHub Actions。

文档

类型	链接
技术文档	文档索引、架构文档、技术报告、源码审计、文件清单
社区规范	贡献指南、安全策略、支持说明、行为准则、变更记录

开源协议

TeXada-the-Math-Agent 使用 GPL-3.0-or-later 协议发布，详见 LICENSE。

Name		Name	Last commit message	Last commit date
Latest commit History 86 Commits
.github		.github
assets		assets
docs		docs
eval		eval
scripts		scripts
src/texada		src/texada
tauri-shell		tauri-shell
tests		tests
.editorconfig		.editorconfig
.env.example		.env.example
.gitattributes		.gitattributes
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
LICENSE		LICENSE
README.md		README.md
package-lock.json		package-lock.json
package.json		package.json
pyproject.toml		pyproject.toml
uv.lock		uv.lock

Folders and files

Latest commit

History

Repository files navigation

TeXada

TeXada

Highlights

Agent Technical Highlights

Download

From A Fresh Machine To First Formula

Quick Start With Ollama

Connection Map And Ports

Cloud Mode

Configuration

Data Backup

Privacy And Limits

Troubleshooting

Experimental CAS Boundary

Release CI

Shortcuts

Hardware And Measurements

Quality Gate

Documentation

License

TeXada

亮点

Agent 技术亮点

下载

从裸机到第一次使用

Ollama 快速启动

连接层级与端口

云侧模式

配置文件

数据备份

隐私与限制

故障排查

实验性 CAS 能力边界

Release CI

快捷键

硬件与实测

质量门禁

文档

开源协议

About

Resources

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Releases

Packages

Used by

Contributors

Languages