lemonmindyes · lemonmindyes · May 11, 2026 · May 11, 2026
diff --git a/README.md b/README.md
@@ -9,7 +9,7 @@ It is designed for researchers, engineers, and students who want to quickly unde
 ## What It Does ✨
 
 - 📄 Analyze academic paper PDFs.
-- 🧠 Generate structured paper summaries with DeepSeek / Jiekou / MiniMax / GLM.
+- 🧠 Generate structured paper summaries with DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo.
 - 🧮 Render Markdown, tables, and LaTeX formulas clearly.
 - 💻 Decide whether the paper needs core code.
 - 🧭 Plan a minimal core-code blueprint before writing files.
@@ -19,7 +19,7 @@ It is designed for researchers, engineers, and students who want to quickly unde
 
 ## Core Workflow 🚀
 
-1. Select a provider (DeepSeek / Jiekou / MiniMax / GLM) and enter your API key in the sidebar.
+1. Select a provider (DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo) and enter your API key in the sidebar.
 2. Choose a model.
 3. Select a paper PDF.
 4. Start analysis.
@@ -50,6 +50,7 @@ Paper2CoreCode supports multiple OpenAI-compatible model providers:
 - Jiekou: Claude, Gemini 3.1 preview, and GPT 5.5 models that work with the current chat completions endpoint.
 - MiniMax: `MiniMax-M2.7`, `MiniMax-M2.7-highspeed`, `MiniMax-M2.5`, `MiniMax-M2.5-highspeed`.
 - GLM: `glm-5.1`, `glm-5`, `glm-5-turbo`.
+- Xiaomi MiMo: `mimo-v2.5-pro`, `mimo-v2-pro`, `mimo-v2.5`.
 
 API keys and model choices are stored separately for each provider in the local app user data directory. Switching providers reloads that provider's own saved key and model.
 
@@ -65,7 +66,7 @@ Version tags like `v0.1.3` trigger the release workflow, which builds platform p
 
 - Electron + TypeScript
 - React + Vite
-- DeepSeek / Jiekou / MiniMax / GLM APIs (OpenAI-compatible)
+- DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo APIs (OpenAI-compatible)
 - `pdf-parse`
 - `react-markdown` + KaTeX
 - `electron-builder`

diff --git a/README.zh-CN.md b/README.zh-CN.md
@@ -9,7 +9,7 @@ Paper2CoreCode 是一款桌面端论文阅读与最小核心代码生成工具
 ## 它能做什么 ✨
 
 - 📄 分析学术论文 PDF。
-- 🧠 使用 DeepSeek / Jiekou / MiniMax / GLM 生成结构化论文总结。
+- 🧠 使用 DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo 生成结构化论文总结。
 - 🧮 清晰渲染 Markdown、表格和 LaTeX 公式。
 - 💻 判断论文是否需要生成核心代码。
 - 🧭 在写入文件前规划最小核心代码蓝图。
@@ -19,7 +19,7 @@ Paper2CoreCode 是一款桌面端论文阅读与最小核心代码生成工具
 
 ## 核心流程 🚀
 
-1. 在侧边栏选择模型供应商（DeepSeek / Jiekou / MiniMax / GLM）并配置 API Key。
+1. 在侧边栏选择模型供应商（DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo）并配置 API Key。
 2. 选择模型。
 3. 选择论文 PDF。
 4. 开始分析。
@@ -50,6 +50,7 @@ Paper2CoreCode 支持多个 OpenAI-compatible 模型供应商：
 - Jiekou：当前 chat completions 流程可用的 Claude、Gemini 3.1 preview 和 GPT 5.5 模型。
 - MiniMax：`MiniMax-M2.7`、`MiniMax-M2.7-highspeed`、`MiniMax-M2.5`、`MiniMax-M2.5-highspeed`。
 - GLM：`glm-5.1`、`glm-5`、`glm-5-turbo`。
+- Xiaomi MiMo：`mimo-v2.5-pro`、`mimo-v2-pro`、`mimo-v2.5`。
 
 API Key 和模型选择会按供应商分别保存在本机应用用户数据目录中。切换供应商时，应用会重新加载该供应商自己的 API Key 和模型。
 
@@ -65,7 +66,7 @@ Pull Request 会通过 GitHub Actions 在 Windows、macOS 和 Linux 上运行 `n
 
 - Electron + TypeScript
 - React + Vite
-- DeepSeek / Jiekou / MiniMax / GLM APIs（OpenAI-compatible）
+- DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo APIs（OpenAI-compatible）
 - `pdf-parse`
 - `react-markdown` + KaTeX
 - `electron-builder`

diff --git a/package.json b/package.json
@@ -1,7 +1,7 @@
 {
   "name": "paper2corecode",
   "version": "0.1.3",
-  "description": "Desktop app to analyze papers and extract core code with DeepSeek / Jiekou / MiniMax / GLM",
+  "description": "Desktop app to analyze papers and extract core code with DeepSeek / Jiekou / MiniMax / GLM / Xiaomi MiMo",
   "author": {
     "name": "lemonmindyes",
     "email": "lemonmindyes@users.noreply.github.com"

diff --git a/src/main/backend/deepseekClient.ts b/src/main/backend/deepseekClient.ts
@@ -18,11 +18,22 @@ interface GlmModelConfig {
   thinking?: { type: 'enabled' }
 }
 
+interface MimoModelConfig {
+  maxTokens?: number
+  tokenParam?: 'max_tokens' | 'max_completion_tokens'
+  temperature?: number
+  topP?: number
+  frequencyPenalty?: number
+  presencePenalty?: number
+  stop?: null
+}
+
 const PROVIDER_CONFIGS: Record<string, ProviderConfig> = {
   deepseek: { baseURL: 'https://api.deepseek.com/v1' },
   jiekou: { baseURL: 'https://api.jiekou.ai/openai' },
   minimax: { baseURL: 'https://api.minimaxi.com/v1' },
   glm: { baseURL: 'https://open.bigmodel.cn/api/paas/v4' },
+  mimo: { baseURL: 'https://api.xiaomimimo.com/v1' },
 }
 
 const JIEKOU_MODEL_CONFIGS: Record<string, JiekouModelConfig> = {
@@ -48,6 +59,36 @@ const GLM_MODEL_CONFIGS: Record<string, GlmModelConfig> = {
   'glm-5-turbo': { maxTokens: 65536, temperature: 1.0, thinking: { type: 'enabled' } },
 }
 
+const MIMO_MODEL_CONFIGS: Record<string, MimoModelConfig> = {
+  'mimo-v2.5-pro': {
+    maxTokens: 131072,
+    tokenParam: 'max_completion_tokens',
+    temperature: 1.0,
+    topP: 0.95,
+    frequencyPenalty: 0,
+    presencePenalty: 0,
+    stop: null,
+  },
+  'mimo-v2-pro': {
+    maxTokens: 131072,
+    tokenParam: 'max_completion_tokens',
+    temperature: 1.0,
+    topP: 0.95,
+    frequencyPenalty: 0,
+    presencePenalty: 0,
+    stop: null,
+  },
+  'mimo-v2.5': {
+    maxTokens: 32768,
+    tokenParam: 'max_completion_tokens',
+    temperature: 1.0,
+    topP: 0.95,
+    frequencyPenalty: 0,
+    presencePenalty: 0,
+    stop: null,
+  },
+}
+
 export function loadConfig() {
   const config = getActiveSettings()
   if (!config.apiKey || config.apiKey.trim() === '') {
@@ -75,7 +116,9 @@ export async function callDeepSeek(
       ? JIEKOU_MODEL_CONFIGS[config.model]
       : config.provider === 'glm'
         ? GLM_MODEL_CONFIGS[config.model]
-        : undefined
+        : config.provider === 'mimo'
+          ? MIMO_MODEL_CONFIGS[config.model]
+          : undefined
 
     if (modelCfg && 'unsupportedReason' in modelCfg && modelCfg.unsupportedReason) {
       throw new AppError(
@@ -92,18 +135,21 @@ export async function callDeepSeek(
       max_tokens?: number
       max_completion_tokens?: number
       temperature?: number
+      top_p?: number
+      frequency_penalty?: number
+      presence_penalty?: number
+      stop?: null
       thinking?: { type: string }
     } = {
       model: config.model,
       messages,
       stream: true,
     }
 
-    if (modelCfg?.maxTokens) {
-      requestBody.max_tokens = modelCfg.maxTokens
-    }
     if (modelCfg && 'tokenParam' in modelCfg && modelCfg.tokenParam && modelCfg.maxTokens) {
       requestBody[modelCfg.tokenParam] = modelCfg.maxTokens
+    } else if (modelCfg?.maxTokens) {
+      requestBody.max_tokens = modelCfg.maxTokens
     }
 
     if (modelCfg && 'temperature' in modelCfg && modelCfg.temperature !== undefined) {
@@ -114,6 +160,22 @@ export async function callDeepSeek(
       requestBody.thinking = modelCfg.thinking
     }
 
+    if (modelCfg && 'topP' in modelCfg && modelCfg.topP !== undefined) {
+      requestBody.top_p = modelCfg.topP
+    }
+
+    if (modelCfg && 'frequencyPenalty' in modelCfg && modelCfg.frequencyPenalty !== undefined) {
+      requestBody.frequency_penalty = modelCfg.frequencyPenalty
+    }
+
+    if (modelCfg && 'presencePenalty' in modelCfg && modelCfg.presencePenalty !== undefined) {
+      requestBody.presence_penalty = modelCfg.presencePenalty
+    }
+
+    if (modelCfg && 'stop' in modelCfg && modelCfg.stop === null) {
+      requestBody.stop = null
+    }
+
     const response = await fetch(`${providerCfg.baseURL}/chat/completions`, {
       method: 'POST',
       headers: {

diff --git a/src/main/backend/settingsStore.ts b/src/main/backend/settingsStore.ts
@@ -34,6 +34,10 @@ export const PROVIDER_SETTINGS: Record<string, { defaultModel: string; models: s
     defaultModel: 'glm-5.1',
     models: ['glm-5.1', 'glm-5', 'glm-5-turbo'],
   },
+  mimo: {
+    defaultModel: 'mimo-v2.5-pro',
+    models: ['mimo-v2.5-pro', 'mimo-v2-pro', 'mimo-v2.5'],
+  },
 }
 
 export interface ActiveSettings {

diff --git a/src/renderer/components/SettingsPanel.tsx b/src/renderer/components/SettingsPanel.tsx
@@ -45,6 +45,15 @@ const PROVIDERS = [
       { value: 'glm-5-turbo', label: 'GLM 5 Turbo' },
     ],
   },
+  {
+    value: 'mimo',
+    label: 'Xiaomi MiMo',
+    models: [
+      { value: 'mimo-v2.5-pro', label: 'MiMo v2.5 Pro' },
+      { value: 'mimo-v2-pro', label: 'MiMo v2 Pro' },
+      { value: 'mimo-v2.5', label: 'MiMo v2.5' },
+    ],
+  },
 ]
 
 const sectionStyle: React.CSSProperties = {