@lobehub/chat 1.118.1 → 1.118.3

package/CHANGELOG.md

@@ -2,6 +2,56 @@
 
 # Changelog
 
+### [Version 1.118.3](https://github.com/lobehub/lobe-chat/compare/v1.118.2...v1.118.3)
+
+<sup>Released on **2025-08-29**</sup>
+
+#### 🐛 Bug Fixes
+
+- **misc**: Correct totalOutputTokens calculation for XAI provider.
+
+<br/>
+
+<details>
+<summary><kbd>Improvements and Fixes</kbd></summary>
+
+#### What's fixed
+
+- **misc**: Correct totalOutputTokens calculation for XAI provider, closes [#8984](https://github.com/lobehub/lobe-chat/issues/8984) ([09ce90a](https://github.com/lobehub/lobe-chat/commit/09ce90a))
+
+</details>
+
+<div align="right">
+
+[![](https://img.shields.io/badge/-BACK_TO_TOP-151515?style=flat-square)](#readme-top)
+
+</div>
+
+### [Version 1.118.2](https://github.com/lobehub/lobe-chat/compare/v1.118.1...v1.118.2)
+
+<sup>Released on **2025-08-29**</sup>
+
+#### 💄 Styles
+
+- **misc**: Add Grok Code Fast 1 model.
+
+<br/>
+
+<details>
+<summary><kbd>Improvements and Fixes</kbd></summary>
+
+#### Styles
+
+- **misc**: Add Grok Code Fast 1 model, closes [#8982](https://github.com/lobehub/lobe-chat/issues/8982) ([dbcec3d](https://github.com/lobehub/lobe-chat/commit/dbcec3d))
+
+</details>
+
+<div align="right">
+
+[![](https://img.shields.io/badge/-BACK_TO_TOP-151515?style=flat-square)](#readme-top)
+
+</div>
+
 ### [Version 1.118.1](https://github.com/lobehub/lobe-chat/compare/v1.118.0...v1.118.1)
 
 <sup>Released on **2025-08-29**</sup>

package/changelog/v1.json

@@ -1,4 +1,22 @@
 [
+  {
+    "children": {
+      "fixes": [
+        "Correct totalOutputTokens calculation for XAI provider."
+      ]
+    },
+    "date": "2025-08-29",
+    "version": "1.118.3"
+  },
+  {
+    "children": {
+      "improvements": [
+        "Add Grok Code Fast 1 model."
+      ]
+    },
+    "date": "2025-08-29",
+    "version": "1.118.2"
+  },
   {
     "children": {
       "fixes": [

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lobehub/chat",
-  "version": "1.118.1",
+  "version": "1.118.3",
   "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",

package/packages/model-runtime/src/ollama/index.ts

@@ -137,8 +137,8 @@ export class LobeOllamaAI implements LobeRuntimeAI {
           enabled: knownModel?.enabled || false,
           functionCall: knownModel?.abilities?.functionCall || false,
           id: model.name,
-          reasoning: knownModel?.abilities?.functionCall || false,
-          vision: knownModel?.abilities?.functionCall || false,
+          reasoning: knownModel?.abilities?.reasoning || false,
+          vision: knownModel?.abilities?.vision || false,
         };
       })
       .filter(Boolean) as ChatModelCard[];

package/packages/model-runtime/src/utils/usageConverter.test.ts

@@ -271,7 +271,7 @@ describe('convertUsage', () => {
     // Assert
     expect(xaiResult).toMatchObject({
       totalInputTokens: 6103,
-      totalOutputTokens: 66,
+      totalOutputTokens: 447, // 66 + 381，xai的reasoning_tokens和completion_tokens价格一样
       outputTextTokens: 66, // 不减去 reasoning_tokens
       outputReasoningTokens: 381,
       totalTokens: 6550,

package/packages/model-runtime/src/utils/usageConverter.ts

@@ -27,6 +27,8 @@ export const convertUsage = (
     provider === 'xai'
       ? totalOutputTokens - outputAudioTokens
       : totalOutputTokens - outputReasoning - outputAudioTokens - outputImageTokens;
+  const totalOutputTokensNormalized =
+    provider === 'xai' ? totalOutputTokens + outputReasoning : totalOutputTokens;
 
   const totalTokens = inputCitationTokens + usage.total_tokens;
 
@@ -43,7 +45,7 @@ export const convertUsage = (
     outputTextTokens: outputTextTokens,
     rejectedPredictionTokens: usage.completion_tokens_details?.rejected_prediction_tokens,
     totalInputTokens,
-    totalOutputTokens: totalOutputTokens,
+    totalOutputTokens: totalOutputTokensNormalized,
     totalTokens,
   } satisfies ModelTokensUsage;
 

package/packages/model-runtime/src/xai/index.ts

@@ -6,7 +6,7 @@ export interface XAIModelCard {
   id: string;
 }
 
-export const GrokReasoningModels = new Set(['grok-3-mini', 'grok-4']);
+export const GrokReasoningModels = new Set(['grok-3-mini', 'grok-4', 'grok-code']);
 
 export const isGrokReasoningModel = (model: string) =>
   Array.from(GrokReasoningModels).some((id) => model.includes(id));

package/packages/types/src/aiModel.ts

@@ -52,6 +52,7 @@ const AiModelAbilitiesSchema = z.object({
   // files: z.boolean().optional(),
   functionCall: z.boolean().optional(),
   reasoning: z.boolean().optional(),
+  search: z.boolean().optional(),
   vision: z.boolean().optional(),
 });
 

package/src/config/aiModels/xai.ts

@@ -2,6 +2,31 @@ import { AIChatModelCard, AIImageModelCard } from '@/types/aiModel';
 
 // https://docs.x.ai/docs/models
 const xaiChatModels: AIChatModelCard[] = [
+  {
+    abilities: {
+      functionCall: true,
+      reasoning: true,
+    },
+    contextWindowTokens: 256_000,
+    description:
+      '我们很高兴推出 grok-code-fast-1，这是一款快速且经济高效的推理模型，在代理编码方面表现出色。',
+    displayName: 'Grok Code Fast 1',
+    enabled: true,
+    id: 'grok-code-fast-1',
+    pricing: {
+      units: [
+        { name: 'textInput_cacheRead', rate: 0.02, strategy: 'fixed', unit: 'millionTokens' },
+        { name: 'textInput', rate: 0.2, strategy: 'fixed', unit: 'millionTokens' },
+        { name: 'textOutput', rate: 1.5, strategy: 'fixed', unit: 'millionTokens' },
+      ],
+    },
+    releasedAt: '2025-08-28',
+    // settings: {
+      // reasoning_effort is not supported by grok-code. Specifying reasoning_effort parameter will get an error response.
+      // extendParams: ['reasoningEffort'],
+    // },
+    type: 'chat',
+  },
   {
     abilities: {
       functionCall: true,