@lobehub/chat 1.65.1 → 1.66.0

package/docker-compose/setup.sh

@@ -394,6 +394,7 @@ SUB_DIR="docker-compose/local"
 FILES=(
     "$SUB_DIR/docker-compose.yml"
     "$SUB_DIR/init_data.json"
+    "$SUB_DIR/searxng-settings.yml"
 )
 ENV_EXAMPLES=(
     "$SUB_DIR/.env.zh-CN.example"
@@ -434,6 +435,7 @@ section_download_files(){
     
     download_file "$SOURCE_URL/${FILES[0]}" "docker-compose.yml"
     download_file "$SOURCE_URL/${FILES[1]}" "init_data.json"
+    download_file "$SOURCE_URL/${FILES[2]}" "searxng-settings.yml"
     
     # Download .env.example with the specified language
     if [ "$LANGUAGE" = "zh_CN" ]; then
@@ -657,4 +659,4 @@ section_display_configurated_report() {
     printf "\n%s" "$(show_message "tips_show_documentation")"
     printf "%s\n" $(show_message "tips_show_documentation_url")
 }
-section_display_configurated_report
+section_display_configurated_report

package/docs/self-hosting/advanced/model-list.mdx

@@ -17,7 +17,7 @@ LobeChat supports customizing the model list during deployment. This configurati
 You can use `+` to add a model, `-` to hide a model, and use `model name=display name<extension configuration>` to customize the display name of a model, separated by English commas. The basic syntax is as follows:
 
 ```text
-id=displayName<maxToken:vision:reasoning:fc:file>,model2,model3
+id=displayName<maxToken:vision:reasoning:search:fc:file>,model2,model3
 ```
 
 For example: `+qwen-7b-chat,+glm-6b,-gpt-3.5-turbo,gpt-4-0125-preview=gpt-4-turbo`
@@ -29,7 +29,7 @@ In the above example, it adds `qwen-7b-chat` and `glm-6b` to the model list, rem
 Considering the diversity of model capabilities, we started to add extension configuration in version `0.147.8`, with the following rules:
 
 ```shell
-id=displayName<maxToken:vision:reasoning:fc:file>
+id=displayName<maxToken:vision:reasoning:search:fc:file>
 ```
 
 The first value in angle brackets is designated as the `maxToken` for this model. The second value and beyond are the model's extension capabilities, separated by colons `:`, and the order is not important.
@@ -40,6 +40,7 @@ Examples are as follows:
 - `spark-v3.5=讯飞星火 v3.5<8192:fc>`: Xunfei Spark 3.5 model, maximum context of 8k, supports Function Call;
 - `gemini-1.5-flash-latest=Gemini 1.5 Flash<16000:vision>`: Google Vision model, maximum context of 16k, supports image recognition;
 - `o3-mini=OpenAI o3-mini<200000:reasoning:fc>`: OpenAI o3-mini model, maximum context of 200k, supports reasoning and Function Call;
+- `qwen-max-latest=Qwen Max<32768:search:fc>`: Qwen 2.5 Max model, maximum context of 32k, supports web search and Function Call;
 - `gpt-4-all=ChatGPT Plus<128000:fc:vision:file>`, hacked version of ChatGPT Plus web, context of 128k, supports image recognition, Function Call, file upload.
 
 Currently supported extension capabilities are:
@@ -49,4 +50,5 @@ Currently supported extension capabilities are:
 | `fc`        | Function Calling                                         |
 | `vision`    | Image Recognition                                        |
 | `reasoning` | Support Reasoning                                        |
+| `search`    | Support Web Search                                        |
 | `file`      | File Upload (a bit hacky, not recommended for daily use) |

package/docs/self-hosting/advanced/model-list.zh-CN.mdx

@@ -16,7 +16,7 @@ LobeChat 支持在部署时自定义模型列表，详情请参考 [模型提供
 你可以使用 `+` 增加一个模型，使用 `-` 来隐藏一个模型，使用 `模型名=展示名<扩展配置>` 来自定义模型的展示名，用英文逗号隔开。通过 `<>` 来添加扩展配置。基本语法如下：
 
 ```text
-id=displayName<maxToken:vision:reasoning:fc:file>,model2,model3
+id=displayName<maxToken:vision:reasoning:search:fc:file>,model2,model3
 ```
 
 例如： `+qwen-7b-chat,+glm-6b,-gpt-3.5-turbo,gpt-4-0125-preview=gpt-4-turbo`
@@ -28,7 +28,7 @@ id=displayName<maxToken:vision:reasoning:fc:file>,model2,model3
 考虑到模型的能力多样性，我们在 `0.147.8` 版本开始增加扩展性配置，它的规则如下：
 
 ```shell
-id=displayName<maxToken:vision:reasoning:fc:file>
+id=displayName<maxToken:vision:reasoning:search:fc:file>
 ```
 
 尖括号第一个值约定为这个模型的 `maxToken` 。第二个及以后作为模型的扩展能力，能力与能力之间用冒号 `:` 作为分隔符，顺序不重要。
@@ -39,6 +39,7 @@ id=displayName<maxToken:vision:reasoning:fc:file>
 - `spark-v3.5=讯飞星火 v3.5<8192:fc>`：讯飞星火 3.5 模型，最大上下文 8k，支持 Function Call；
 - `gemini-1.5-flash-latest=Gemini 1.5 Flash<16000:vision>`：Google 视觉模型，最大上下文 16k，支持图像识别；
 - `o3-mini=OpenAI o3-mini<200000:reasoning:fc>`：OpenAI o3-mini 模型，最大上下文 200k，支持推理及 Function Call；
+- `qwen-max-latest=Qwen Max<32768:search:fc>`：通义千问 2.5 Max 模型，最大上下文 32k，支持联网搜索及 Function Call；
 - `gpt-4-all=ChatGPT Plus<128000:fc:vision:file>`，hack 的 ChatGPT Plus 网页版，上下 128k ，支持图像识别、Function Call、文件上传
 
 目前支持的扩展能力有：
@@ -48,4 +49,5 @@ id=displayName<maxToken:vision:reasoning:fc:file>
 | `fc`        | 函数调用（function calling） |
 | `vision`    | 视觉识别                   |
 | `reasoning` | 支持推理                   |
+| `search`    | 支持联网搜索                   |
 | `file`      | 文件上传（比较 hack，不建议日常使用）  |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lobehub/chat",
-  "version": "1.65.1",
+  "version": "1.66.0",
   "description": "Lobe Chat - an open-source, high-performance chatbot framework that supports speech synthesis, multimodal, and extensible Function Call plugin system. Supports one-click free deployment of your private ChatGPT/LLM web application.",
   "keywords": [
     "framework",
@@ -131,7 +131,7 @@
     "@lobehub/tts": "^1.28.0",
     "@lobehub/ui": "^1.165.2",
     "@neondatabase/serverless": "^0.10.4",
-    "@next/third-parties": "^15.1.4",
+    "@next/third-parties": "^15.2.0",
     "@react-spring/web": "^9.7.5",
     "@sentry/nextjs": "^7.120.2",
     "@serwist/next": "^9.0.11",
@@ -179,7 +179,7 @@
     "mdast-util-to-markdown": "^2.1.2",
     "modern-screenshot": "^4.5.5",
     "nanoid": "^5.0.9",
-    "next": "^15.1.4",
+    "next": "^15.2.0",
     "next-auth": "beta",
     "next-mdx-remote": "^5.0.0",
     "nextjs-toploader": "^3.7.15",
@@ -250,8 +250,8 @@
     "@lobehub/i18n-cli": "^1.20.3",
     "@lobehub/lint": "^1.25.5",
     "@lobehub/seo-cli": "^1.4.3",
-    "@next/bundle-analyzer": "^15.1.4",
-    "@next/eslint-plugin-next": "^15.1.4",
+    "@next/bundle-analyzer": "^15.2.0",
+    "@next/eslint-plugin-next": "^15.2.0",
     "@peculiar/webcrypto": "^1.5.0",
     "@semantic-release/exec": "^6.0.3",
     "@testing-library/jest-dom": "^6.6.3",
@@ -269,8 +269,8 @@
     "@types/node": "^22.10.5",
     "@types/numeral": "^2.0.5",
     "@types/pg": "^8.11.10",
-    "@types/react": "^19.0.3",
-    "@types/react-dom": "^19.0.2",
+    "@types/react": "^19.0.10",
+    "@types/react-dom": "^19.0.4",
     "@types/rtl-detect": "^1.0.3",
     "@types/semver": "^7.5.8",
     "@types/systemjs": "^6.15.1",

package/src/config/aiModels/ai360.ts

@@ -52,6 +52,7 @@ const ai360ChatModels: AIChatModelCard[] = [
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 8000,
     description:
@@ -64,6 +65,9 @@ const ai360ChatModels: AIChatModelCard[] = [
       input: 2,
       output: 5,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {

package/src/config/aiModels/baichuan.ts

@@ -4,6 +4,7 @@ const baichuanChatModels: AIChatModelCard[] = [
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_768,
     description:
@@ -17,11 +18,15 @@ const baichuanChatModels: AIChatModelCard[] = [
       input: 100,
       output: 100,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_768,
     description:
@@ -35,11 +40,15 @@ const baichuanChatModels: AIChatModelCard[] = [
       input: 15,
       output: 15,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_768,
     description:
@@ -53,11 +62,15 @@ const baichuanChatModels: AIChatModelCard[] = [
       input: 0.98,
       output: 0.98,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_768,
     description:
@@ -70,6 +83,9 @@ const baichuanChatModels: AIChatModelCard[] = [
       input: 12,
       output: 12,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {

package/src/config/aiModels/google.ts

@@ -80,9 +80,9 @@ const googleChatModels: AIChatModelCard[] = [
       vision: true,
     },
     contextWindowTokens: 1_048_576 + 8192,
-    description: '一个 Gemini 2.0 Flash 模型，针对成本效益和低延迟等目标进行了优化。',
-    displayName: 'Gemini 2.0 Flash-Lite Preview 02-05',
-    id: 'gemini-2.0-flash-lite-preview-02-05',
+    description: 'Gemini 2.0 Flash 模型变体，针对成本效益和低延迟等目标进行了优化。',
+    displayName: 'Gemini 2.0 Flash-Lite 001',
+    id: 'gemini-2.0-flash-lite-001',
     maxOutput: 8192,
     pricing: {
       cachedInput: 0.018_75,

package/src/config/aiModels/groq.ts

@@ -176,6 +176,16 @@ const groqChatModels: AIChatModelCard[] = [
     },
     type: 'chat',
   },
+  {
+    contextWindowTokens: 32_768,
+    displayName: 'Mixtral Saba 24B',
+    id: 'mistral-saba-24b',
+    pricing: {
+      input: 0.79,
+      output: 0.79,
+    },
+    type: 'chat',
+  },
   {
     abilities: {
       functionCall: true,

package/src/config/aiModels/hunyuan.ts

@@ -20,6 +20,9 @@ const hunyuanChatModels: AIChatModelCard[] = [
     type: 'chat',
   },
   {
+    abilities: {
+      search: true,
+    },
     contextWindowTokens: 32_000,
     description:
       '采用更优的路由策略，同时缓解了负载均衡和专家趋同的问题。长文方面，大海捞针指标达到99.9%。MOE-32K 性价比相对更高，在平衡效果、价格的同时，可对实现对长文本输入的处理。',
@@ -33,9 +36,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 2,
     },
     releasedAt: '2025-02-10',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
+    abilities: {
+      search: true,
+    },
     contextWindowTokens: 256_000,
     description:
       '采用更优的路由策略，同时缓解了负载均衡和专家趋同的问题。长文方面，大海捞针指标达到99.9%。MOE-256K 在长度和效果上进一步突破，极大的扩展了可输入长度。',
@@ -49,11 +58,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 2,
     },
     releasedAt: '2025-02-10',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_000,
     description:
@@ -68,11 +81,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 50,
     },
     releasedAt: '2025-01-10',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_000,
     description:
@@ -86,11 +103,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 50,
     },
     releasedAt: '2025-01-10',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_000,
     description:
@@ -104,11 +125,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 50,
     },
     releasedAt: '2025-01-10',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_000,
     description:
@@ -122,9 +147,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 50,
     },
     releasedAt: '2024-11-20',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
+    abilities: {
+      search: true,
+    },
     contextWindowTokens: 32_000,
     description:
       'Hunyuan-large 模型总参数量约 389B，激活参数量约 52B，是当前业界参数规模最大、效果最好的 Transformer 架构的开源 MoE 模型。',
@@ -138,9 +169,15 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 12,
     },
     releasedAt: '2025-02-10',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
+    abilities: {
+      search: true,
+    },
     contextWindowTokens: 134_000,
     description:
       '擅长处理长文任务如文档摘要和文档问答等，同时也具备处理通用文本生成任务的能力。在长文本的分析和生成上表现优异，能有效应对复杂和详尽的长文内容处理需求。',
@@ -154,6 +191,9 @@ const hunyuanChatModels: AIChatModelCard[] = [
       output: 18,
     },
     releasedAt: '2024-12-18',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {

package/src/config/aiModels/moonshot.ts

@@ -4,6 +4,7 @@ const moonshotChatModels: AIChatModelCard[] = [
   {
     abilities: {
       functionCall: true,
+      search: true,
       vision: true,
     },
     contextWindowTokens: 131_072,
@@ -19,11 +20,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       output: 60,
     },
     releasedAt: '2025-02-17',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 131_072,
     description:
@@ -35,11 +40,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       input: 60,
       output: 60,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 8192,
     description:
@@ -51,11 +60,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       input: 12,
       output: 12,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 32_768,
     description:
@@ -67,11 +80,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       input: 24,
       output: 24,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
     },
     contextWindowTokens: 131_072,
     description:
@@ -83,11 +100,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       input: 60,
       output: 60,
     },
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
       vision: true,
     },
     contextWindowTokens: 8192,
@@ -101,11 +122,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       output: 12,
     },
     releasedAt: '2025-01-14',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
       vision: true,
     },
     contextWindowTokens: 32_768,
@@ -119,11 +144,15 @@ const moonshotChatModels: AIChatModelCard[] = [
       output: 24,
     },
     releasedAt: '2025-01-14',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
   {
     abilities: {
       functionCall: true,
+      search: true,
       vision: true,
     },
     contextWindowTokens: 131_072,
@@ -137,6 +166,9 @@ const moonshotChatModels: AIChatModelCard[] = [
       output: 60,
     },
     releasedAt: '2025-01-14',
+    settings: {
+      searchImpl: 'params',
+    },
     type: 'chat',
   },
 ];

package/src/config/aiModels/qwen.ts

@@ -62,8 +62,8 @@ const qwenChatModels: AIChatModelCard[] = [
     organization: 'Qwen',
     pricing: {
       currency: 'CNY',
-      input: 20,
-      output: 60,
+      input: 2.4,
+      output: 9.6,
     },
     settings: {
       searchImpl: 'params',
@@ -410,6 +410,23 @@ const qwenChatModels: AIChatModelCard[] = [
     },
     type: 'chat',
   },
+  {
+    abilities: {
+      vision: true,
+    },
+    contextWindowTokens: 32_768,
+    description: 'Qwen-Omni 系列模型支持输入多种模态的数据，包括视频、音频、图片、文本，并输出音频与文本。',
+    displayName: 'Qwen Omni Turbo',
+    id: 'qwen-omni-turbo-latest',
+    maxOutput: 2048,
+    organization: 'Qwen',
+    pricing: {
+      currency: 'CNY',
+      input: 0,
+      output: 0,
+    },
+    type: 'chat',
+  },
   {
     abilities: {
       vision: true,
@@ -452,7 +469,7 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 65_792,
     description:
       'DeepSeek-R1 在后训练阶段大规模使用了强化学习技术，在仅有极少标注数据的情况下，极大提升了模型推理能力。在数学、代码、自然语言推理等任务上，性能较高，能力较强。',
     displayName: 'DeepSeek R1',
@@ -462,8 +479,8 @@ const qwenChatModels: AIChatModelCard[] = [
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
-      input: 0,
-      output: 0,
+      input: 4,
+      output: 16,
     },
     releasedAt: '2025-01-27',
     type: 'chat',
@@ -472,7 +489,7 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       functionCall: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 65_792,
     description:
       'DeepSeek-V3 为自研 MoE 模型，671B 参数，激活 37B，在 14.8T token 上进行了预训练，在长文本、代码、数学、百科、中文能力上表现优秀。',
     displayName: 'DeepSeek V3',
@@ -482,8 +499,8 @@ const qwenChatModels: AIChatModelCard[] = [
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
-      input: 0,
-      output: 0,
+      input: 2,
+      output: 8,
     },
     releasedAt: '2025-01-27',
     type: 'chat',
@@ -492,12 +509,12 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 32_768,
     description:
       'DeepSeek-R1-Distill-Qwen-1.5B 是一个基于 Qwen2.5-Math-1.5B 的蒸馏大型语言模型，使用了 DeepSeek R1 的输出。',
     displayName: 'DeepSeek R1 Distill Qwen 1.5B',
     id: 'deepseek-r1-distill-qwen-1.5b',
-    maxOutput: 8192,
+    maxOutput: 16_384,
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
@@ -510,17 +527,17 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 32_768,
     description:
       'DeepSeek-R1-Distill-Qwen-7B 是一个基于 Qwen2.5-Math-7B 的蒸馏大型语言模型，使用了 DeepSeek R1 的输出。',
     displayName: 'DeepSeek R1 Distill Qwen 7B',
     id: 'deepseek-r1-distill-qwen-7b',
-    maxOutput: 8192,
+    maxOutput: 16_384,
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
-      input: 0,
-      output: 0,
+      input: 0.5,
+      output: 1,
     },
     type: 'chat',
   },
@@ -528,17 +545,17 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 32_768,
     description:
       'DeepSeek-R1-Distill-Qwen-14B 是一个基于 Qwen2.5-14B 的蒸馏大型语言模型，使用了 DeepSeek R1 的输出。',
     displayName: 'DeepSeek R1 Distill Qwen 14B',
     id: 'deepseek-r1-distill-qwen-14b',
-    maxOutput: 8192,
+    maxOutput: 16_384,
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
-      input: 0,
-      output: 0,
+      input: 1,
+      output: 3,
     },
     type: 'chat',
   },
@@ -546,17 +563,17 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 32_768,
     description:
       'DeepSeek-R1-Distill-Qwen-32B 是一个基于 Qwen2.5-32B 的蒸馏大型语言模型，使用了 DeepSeek R1 的输出。',
     displayName: 'DeepSeek R1 Distill Qwen 32B',
     id: 'deepseek-r1-distill-qwen-32b',
-    maxOutput: 8192,
+    maxOutput: 16_384,
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
-      input: 0,
-      output: 0,
+      input: 2,
+      output: 6,
     },
     type: 'chat',
   },
@@ -564,12 +581,12 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 32_768,
     description:
       'DeepSeek-R1-Distill-Llama-8B 是一个基于 Llama-3.1-8B 的蒸馏大型语言模型，使用了 DeepSeek R1 的输出。',
     displayName: 'DeepSeek R1 Distill Llama 8B',
     id: 'deepseek-r1-distill-llama-8b',
-    maxOutput: 8192,
+    maxOutput: 16_384,
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',
@@ -582,12 +599,12 @@ const qwenChatModels: AIChatModelCard[] = [
     abilities: {
       reasoning: true,
     },
-    contextWindowTokens: 131_072,
+    contextWindowTokens: 32_768,
     description:
       'DeepSeek-R1-Distill-Llama-70B 是一个基于 Llama-3.3-70B-Instruct 的蒸馏大型语言模型，使用了 DeepSeek R1 的输出。',
     displayName: 'DeepSeek R1 Distill Llama 70B',
     id: 'deepseek-r1-distill-llama-70b',
-    maxOutput: 8192,
+    maxOutput: 16_384,
     organization: 'DeepSeek',
     pricing: {
       currency: 'CNY',