Release/v0.4.9 (#2421)

* fix: turn off experimental settings should also turn off quick ask (#2411)

* fix: app glitches 1s generating response before starting model (#2412)

* fix: disable experimental feature should also disable vulkan (#2414)

* fix: model load stuck on windows when can't get CPU core count (#2413)

Signed-off-by: James <james@jan.ai>
Co-authored-by: James <james@jan.ai>

* feat: TensorRT-LLM engine update support (#2415)

* fix: engine update

* chore: add remove prepopulated models

Signed-off-by: James <james@jan.ai>

* update tinyjensen url

Signed-off-by: James <james@jan.ai>

* update llamacorn

Signed-off-by: James <james@jan.ai>

* update Mistral 7B Instruct v0.1 int4

Signed-off-by: James <james@jan.ai>

* update tensorrt

Signed-off-by: James <james@jan.ai>

* update

Signed-off-by: hiro <hiro@jan.ai>

* update

Signed-off-by: James <james@jan.ai>

* prettier

Signed-off-by: James <james@jan.ai>

* update mistral config

Signed-off-by: James <james@jan.ai>

* fix some lint

Signed-off-by: James <james@jan.ai>

---------

Signed-off-by: James <james@jan.ai>
Signed-off-by: hiro <hiro@jan.ai>
Co-authored-by: James <james@jan.ai>
Co-authored-by: hiro <hiro@jan.ai>

* Tensorrt LLM disable turing support (#2418)

Co-authored-by: Hien To <tominhhien97@gmail.com>

* chore: add prompt template tensorrtllm (#2375)

* chore: add prompt template tensorrtllm

* Add Prompt template for mistral and correct model metadata

---------

Co-authored-by: Hien To <tominhhien97@gmail.com>

* fix: correct tensorrt mistral model.json (#2419)

---------

Signed-off-by: James <james@jan.ai>
Signed-off-by: hiro <hiro@jan.ai>
Co-authored-by: Louis <louis@jan.ai>
Co-authored-by: James <james@jan.ai>
Co-authored-by: hiro <hiro@jan.ai>
Co-authored-by: hiento09 <136591877+hiento09@users.noreply.github.com>
Co-authored-by: Hien To <tominhhien97@gmail.com>

.github/workflows/jan-electron-linter-and-test.yml

@@ -22,6 +22,7 @@ on:
     branches:
       - main
       - dev
+      - release/**
     paths:
       - "electron/**"
       - .github/workflows/jan-electron-linter-and-test.yml

core/package.json

@@ -46,7 +46,7 @@
   },
   "devDependencies": {
     "@types/jest": "^29.5.12",
-    "@types/node": "^12.0.2",
+    "@types/node": "^20.11.4",
     "eslint": "8.57.0",
     "eslint-plugin-jest": "^27.9.0",
     "jest": "^29.7.0",

core/src/api/index.ts

@@ -96,6 +96,7 @@ export enum FileManagerRoute {
   fileStat = 'fileStat',
   writeBlob = 'writeBlob',
   mkdir = 'mkdir',
+  rm = 'rm',
 }
 
 export type ApiFunction = (...args: any[]) => any

core/src/extension.ts

@@ -19,6 +19,7 @@ export interface Compatibility {
 const ALL_INSTALLATION_STATE = [
   'NotRequired', // not required.
   'Installed', // require and installed. Good to go.
+  'Updatable', // require and installed but need to be updated.
   'NotInstalled', // require to be installed.
   'Corrupted', // require but corrupted. Need to redownload.
 ] as const
@@ -59,6 +60,13 @@ export abstract class BaseExtension implements ExtensionType {
     return undefined
   }
 
+  /**
+   * Determine if the extension is updatable.
+   */
+  updatable(): boolean {
+    return false
+  }
+
   /**
    * Determine if the prerequisites for the extension are installed.
    *

core/src/fs.ts

@@ -45,6 +45,9 @@ const mkdir = (...args: any[]) => global.core.api?.mkdir(...args)
  */
 const rmdirSync = (...args: any[]) =>
   global.core.api?.rmdirSync(...args, { recursive: true, force: true })
+
+const rm = (path: string) => global.core.api?.rm(path)
+
 /**
  * Deletes a file from the local file system.
  * @param {string} path - The path of the file to delete.
@@ -96,6 +99,7 @@ export const fs = {
   mkdirSync,
   mkdir,
   rmdirSync,
+  rm,
   unlinkSync,
   appendFileSync,
   copyFileSync,

core/src/node/api/processors/fsExt.ts

@@ -100,4 +100,16 @@ export class FSExt implements Processor {
       })
     })
   }
+
+  rmdir(path: string): Promise<void> {
+    return new Promise((resolve, reject) => {
+      fs.rm(path, { recursive: true }, (err) => {
+        if (err) {
+          reject(err)
+        } else {
+          resolve()
+        }
+      })
+    })
+  }
 }

core/src/node/helper/config.ts

@@ -82,26 +82,34 @@ export const getJanExtensionsPath = (): string => {
  */
 export const physicalCpuCount = async (): Promise<number> => {
   const platform = os.platform()
-  if (platform === 'linux') {
+  try {
-    const output = await exec('lscpu -p | egrep -v "^#" | sort -u -t, -k 2,4 | wc -l')
+    if (platform === 'linux') {
-    return parseInt(output.trim(), 10)
+      const output = await exec('lscpu -p | egrep -v "^#" | sort -u -t, -k 2,4 | wc -l')
-  } else if (platform === 'darwin') {
+      return parseInt(output.trim(), 10)
-    const output = await exec('sysctl -n hw.physicalcpu_max')
+    } else if (platform === 'darwin') {
-    return parseInt(output.trim(), 10)
+      const output = await exec('sysctl -n hw.physicalcpu_max')
-  } else if (platform === 'win32') {
+      return parseInt(output.trim(), 10)
-    const output = await exec('WMIC CPU Get NumberOfCores')
+    } else if (platform === 'win32') {
-    return output
+      const output = await exec('WMIC CPU Get NumberOfCores')
-      .split(os.EOL)
+      return output
-      .map((line: string) => parseInt(line))
+        .split(os.EOL)
-      .filter((value: number) => !isNaN(value))
+        .map((line: string) => parseInt(line))
-      .reduce((sum: number, number: number) => sum + number, 1)
+        .filter((value: number) => !isNaN(value))
-  } else {
+        .reduce((sum: number, number: number) => sum + number, 1)
-    const cores = os.cpus().filter((cpu: any, index: number) => {
+    } else {
-      const hasHyperthreading = cpu.model.includes('Intel')
+      const cores = os.cpus().filter((cpu: any, index: number) => {
-      const isOdd = index % 2 === 1
+        const hasHyperthreading = cpu.model.includes('Intel')
-      return !hasHyperthreading || isOdd
+        const isOdd = index % 2 === 1
-    })
+        return !hasHyperthreading || isOdd
-    return cores.length
+      })
+      return cores.length
+    }
+  } catch (err) {
+    console.warn('Failed to get physical CPU count', err)
+    // Divide by 2 to get rid of hyper threading
+    const coreCount = Math.ceil(os.cpus().length / 2)
+    console.debug('Using node API to get physical CPU count:', coreCount)
+    return coreCount
   }
 }
 

core/src/node/helper/resource.ts

@@ -1,6 +1,6 @@
 import { SystemResourceInfo } from '../../types'
 import { physicalCpuCount } from './config'
-import { log, logServer } from './log'
+import { log } from './log'
 
 export const getSystemResourceInfo = async (): Promise<SystemResourceInfo> => {
   const cpu = await physicalCpuCount()

extensions/model-extension/src/index.ts

@@ -38,7 +38,7 @@ export default class JanModelExtension extends ModelExtension {
   private static readonly _tensorRtEngineFormat = '.engine'
   private static readonly _configDirName = 'config'
   private static readonly _defaultModelFileName = 'default-model.json'
-  private static readonly _supportedGpuArch = ['turing', 'ampere', 'ada']
+  private static readonly _supportedGpuArch = ['ampere', 'ada']
 
   /**
    * Called when the extension is loaded.

extensions/monitoring-extension/src/node/index.ts

@@ -181,8 +181,7 @@ const updateNvidiaDriverInfo = async () =>
 const getGpuArch = (gpuName: string): string => {
   if (!gpuName.toLowerCase().includes('nvidia')) return 'unknown'
 
-  if (gpuName.includes('20')) return 'turing'
+  if (gpuName.includes('30')) return 'ampere'
-  else if (gpuName.includes('30')) return 'ampere'
   else if (gpuName.includes('40')) return 'ada'
   else return 'unknown'
 }

extensions/tensorrt-llm-extension/models.json

@@ -3,27 +3,31 @@
     "sources": [
       {
         "filename": "config.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/LlamaCorn-1.1B-Chat-fp16/config.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/config.json"
       },
       {
-        "filename": "rank0.engine",
+        "filename": "mistral_float16_tp1_rank0.engine",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/LlamaCorn-1.1B-Chat-fp16/rank0.engine"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/mistral_float16_tp1_rank0.engine"
       },
       {
         "filename": "tokenizer.model",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/LlamaCorn-1.1B-Chat-fp16/tokenizer.model"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/tokenizer.model"
       },
       {
         "filename": "special_tokens_map.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/LlamaCorn-1.1B-Chat-fp16/special_tokens_map.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/special_tokens_map.json"
       },
       {
         "filename": "tokenizer.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/LlamaCorn-1.1B-Chat-fp16/tokenizer.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/tokenizer.json"
       },
       {
         "filename": "tokenizer_config.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/LlamaCorn-1.1B-Chat-fp16/tokenizer_config.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/tokenizer_config.json"
+      },
+      {
+        "filename": "model.cache",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/LlamaCorn-1.1B-Chat-fp16/model.cache"
       }
     ],
     "id": "llamacorn-1.1b-chat-fp16",
@@ -50,27 +54,31 @@
     "sources": [
       {
         "filename": "config.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/TinyJensen-1.1B-Chat-fp16/config.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/config.json"
       },
       {
-        "filename": "rank0.engine",
+        "filename": "mistral_float16_tp1_rank0.engine",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/TinyJensen-1.1B-Chat-fp16/rank0.engine"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/mistral_float16_tp1_rank0.engine"
       },
       {
         "filename": "tokenizer.model",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/TinyJensen-1.1B-Chat-fp16/tokenizer.model"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/tokenizer.model"
       },
       {
         "filename": "special_tokens_map.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/TinyJensen-1.1B-Chat-fp16/special_tokens_map.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/special_tokens_map.json"
       },
       {
         "filename": "tokenizer.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/TinyJensen-1.1B-Chat-fp16/tokenizer.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/tokenizer.json"
       },
       {
         "filename": "tokenizer_config.json",
-        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/TinyJensen-1.1B-Chat-fp16/tokenizer_config.json"
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/tokenizer_config.json"
+      },
+      {
+        "filename": "model.cache",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/TinyJensen-1.1B-Chat-fp16/model.cache"
       }
     ],
     "id": "tinyjensen-1.1b-chat-fp16",
@@ -92,5 +100,57 @@
       "size": 2151000000
     },
     "engine": "nitro-tensorrt-llm"
+  },
+  {
+    "sources": [
+      {
+        "filename": "config.json",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/config.json"
+      },
+      {
+        "filename": "mistral_float16_tp1_rank0.engine",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/mistral_float16_tp1_rank0.engine"
+      },
+      {
+        "filename": "tokenizer.model",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/tokenizer.model"
+      },
+      {
+        "filename": "special_tokens_map.json",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/special_tokens_map.json"
+      },
+      {
+        "filename": "tokenizer.json",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/tokenizer.json"
+      },
+      {
+        "filename": "tokenizer_config.json",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/tokenizer_config.json"
+      },
+      {
+        "filename": "model.cache",
+        "url": "https://delta.jan.ai/dist/models/<gpuarch>/<os>/tensorrt-llm-v0.7.1/Mistral-7B-Instruct-v0.1-int4/model.cache"
+      }
+    ],
+    "id": "mistral-7b-instruct-int4",
+    "object": "model",
+    "name": "Mistral 7B Instruct v0.1 INT4",
+    "version": "1.0",
+    "description": "Mistral 7B Instruct v0.1 INT4",
+    "format": "TensorRT-LLM",
+    "settings": {
+      "ctx_len": 2048,
+      "text_model": false,
+      "prompt_template": "[INST] {prompt} [/INST]"
+    },
+    "parameters": {
+      "max_tokens": 4096
+    },
+    "metadata": {
+      "author": "MistralAI",
+      "tags": ["TensorRT-LLM", "7B", "Finetuned"],
+      "size": 3840000000
+    },
+    "engine": "nitro-tensorrt-llm"
   }
 ]

extensions/tensorrt-llm-extension/package.json

@@ -18,7 +18,7 @@
       "0.1.0"
     ]
   },
-  "tensorrtVersion": "0.1.6",
+  "tensorrtVersion": "0.1.8",
   "provider": "nitro-tensorrt-llm",
   "scripts": {
     "build": "tsc --module commonjs && rollup -c rollup.config.ts",

extensions/tensorrt-llm-extension/rollup.config.ts

@@ -21,7 +21,7 @@ export default [
         DOWNLOAD_RUNNER_URL:
           process.platform === 'win32'
             ? JSON.stringify(
-                'https://github.com/janhq/nitro-tensorrt-llm/releases/download/windows-v<version>/nitro-windows-v<version>-amd64-tensorrt-llm-<gpuarch>.tar.gz'
+                'https://github.com/janhq/nitro-tensorrt-llm/releases/download/windows-v<version>-tensorrt-llm-v0.7.1/nitro-windows-v<version>-tensorrt-llm-v0.7.1-amd64-all-arch.tar.gz'
               )
             : JSON.stringify(
                 'https://github.com/janhq/nitro-tensorrt-llm/releases/download/linux-v<version>/nitro-linux-v<version>-amd64-tensorrt-llm-<gpuarch>.tar.gz'

extensions/tensorrt-llm-extension/src/index.ts

@@ -39,8 +39,9 @@ export default class TensorRTLLMExtension extends LocalOAIEngine {
   override inferenceUrl = INFERENCE_URL
   override nodeModule = NODE
 
-  private supportedGpuArch = ['turing', 'ampere', 'ada']
+  private supportedGpuArch = ['ampere', 'ada']
   private supportedPlatform = ['win32', 'linux']
+  private isUpdateAvailable = false
 
   compatibility() {
     return COMPATIBILITY as unknown as Compatibility
@@ -56,6 +57,8 @@ export default class TensorRTLLMExtension extends LocalOAIEngine {
   }
 
   override async install(): Promise<void> {
+    await this.removePopulatedModels()
+
     const info = await systemInformation()
     console.debug(
       `TensorRTLLMExtension installing pre-requisites... ${JSON.stringify(info)}`
@@ -141,6 +144,22 @@ export default class TensorRTLLMExtension extends LocalOAIEngine {
     events.on(DownloadEvent.onFileDownloadSuccess, onFileDownloadSuccess)
   }
 
+  async removePopulatedModels(): Promise<void> {
+    console.debug(`removePopulatedModels`, JSON.stringify(models))
+    const janDataFolderPath = await getJanDataFolderPath()
+    const modelFolderPath = await joinPath([janDataFolderPath, 'models'])
+
+    for (const model of models) {
+      const modelPath = await joinPath([modelFolderPath, model.id])
+      console.debug(`modelPath: ${modelPath}`)
+      if (await fs.existsSync(modelPath)) {
+        console.debug(`Removing model ${modelPath}`)
+        await fs.rmdirSync(modelPath)
+      }
+    }
+    events.emit(ModelEvent.OnModelsUpdate, {})
+  }
+
   async onModelInit(model: Model): Promise<void> {
     if (model.engine !== this.provider) return
 
@@ -156,6 +175,10 @@ export default class TensorRTLLMExtension extends LocalOAIEngine {
     }
   }
 
+  updatable() {
+    return this.isUpdateAvailable
+  }
+
   override async installationState(): Promise<InstallationState> {
     const info = await systemInformation()
 

extensions/tensorrt-llm-extension/src/node/index.ts

@@ -5,12 +5,13 @@ import fetchRT from 'fetch-retry'
 import { log, getJanDataFolderPath } from '@janhq/core/node'
 import decompress from 'decompress'
 import { SystemInformation } from '@janhq/core'
+import { PromptTemplate } from '@janhq/core'
 
 // Polyfill fetch with retry
 const fetchRetry = fetchRT(fetch)
 
 const supportedPlatform = (): string[] => ['win32', 'linux']
-const supportedGpuArch = (): string[] => ['turing', 'ampere', 'ada']
+const supportedGpuArch = (): string[] => ['ampere', 'ada']
 
 /**
  * The response object for model init operation.
@@ -35,9 +36,21 @@ async function loadModel(
   // e.g. ~/jan/models/llama-2
   let modelFolder = params.modelFolder
 
+  if (params.model.settings.prompt_template) {
+    const promptTemplate = params.model.settings.prompt_template
+    const prompt = promptTemplateConverter(promptTemplate)
+    if (prompt?.error) {
+      return Promise.reject(prompt.error)
+    }
+    params.model.settings.system_prompt = prompt.system_prompt
+    params.model.settings.user_prompt = prompt.user_prompt
+    params.model.settings.ai_prompt = prompt.ai_prompt
+  }
+
   const settings: ModelLoadParams = {
     engine_path: modelFolder,
     ctx_len: params.model.settings.ctx_len ?? 2048,
+    ...params.model.settings,
   }
   if (!systemInfo) {
     throw new Error('Cannot get system info. Unable to start nitro x tensorrt.')
@@ -220,6 +233,52 @@ const decompressRunner = async (zipPath: string, output: string) => {
   }
 }
 
+/**
+ * Parse prompt template into agrs settings
+ * @param promptTemplate Template as string
+ * @returns
+ */
+function promptTemplateConverter(promptTemplate: string): PromptTemplate {
+  // Split the string using the markers
+  const systemMarker = '{system_message}'
+  const promptMarker = '{prompt}'
+
+  if (
+    promptTemplate.includes(systemMarker) &&
+    promptTemplate.includes(promptMarker)
+  ) {
+    // Find the indices of the markers
+    const systemIndex = promptTemplate.indexOf(systemMarker)
+    const promptIndex = promptTemplate.indexOf(promptMarker)
+
+    // Extract the parts of the string
+    const system_prompt = promptTemplate.substring(0, systemIndex)
+    const user_prompt = promptTemplate.substring(
+      systemIndex + systemMarker.length,
+      promptIndex
+    )
+    const ai_prompt = promptTemplate.substring(
+      promptIndex + promptMarker.length
+    )
+
+    // Return the split parts
+    return { system_prompt, user_prompt, ai_prompt }
+  } else if (promptTemplate.includes(promptMarker)) {
+    // Extract the parts of the string for the case where only promptMarker is present
+    const promptIndex = promptTemplate.indexOf(promptMarker)
+    const user_prompt = promptTemplate.substring(0, promptIndex)
+    const ai_prompt = promptTemplate.substring(
+      promptIndex + promptMarker.length
+    )
+
+    // Return the split parts
+    return { user_prompt, ai_prompt }
+  }
+
+  // Return an error if none of the conditions are met
+  return { error: 'Cannot split prompt template' }
+}
+
 export default {
   supportedPlatform,
   supportedGpuArch,

web/containers/Providers/index.tsx

@@ -1,6 +1,6 @@
 'use client'
 
-import { PropsWithChildren, useEffect, useState } from 'react'
+import { PropsWithChildren, useCallback, useEffect, useState } from 'react'
 
 import { Toaster } from 'react-hot-toast'
 
@@ -37,7 +37,7 @@ const Providers = (props: PropsWithChildren) => {
   const [activated, setActivated] = useState(false)
   const [settingUp, setSettingUp] = useState(false)
 
-  async function setupExtensions() {
+  const setupExtensions = useCallback(async () => {
     // Register all active extensions
     await extensionManager.registerActive()
 
@@ -57,7 +57,7 @@ const Providers = (props: PropsWithChildren) => {
       setSettingUp(false)
       setActivated(true)
     }, 500)
-  }
+  }, [pathname])
 
   // Services Setup
   useEffect(() => {
@@ -78,7 +78,7 @@ const Providers = (props: PropsWithChildren) => {
         setActivated(true)
       }
     }
-  }, [setupCore])
+  }, [setupCore, setupExtensions])
 
   return (
     <JotaiWrapper>

web/hooks/useSendChatMessage.ts

@@ -102,7 +102,6 @@ export default function useSendChatMessage() {
       console.error('No active thread')
       return
     }
-    setIsGeneratingResponse(true)
     updateThreadWaiting(activeThreadRef.current.id, true)
     const messages: ChatCompletionMessage[] = [
       activeThreadRef.current.assistants[0]?.instructions,
@@ -148,7 +147,7 @@ export default function useSendChatMessage() {
       await waitForModelStarting(modelId)
       setQueuedMessage(false)
     }
-
+    setIsGeneratingResponse(true)
     if (currentMessage.role !== ChatCompletionRole.User) {
       // Delete last response before regenerating
       deleteMessage(currentMessage.id ?? '')
@@ -171,7 +170,6 @@ export default function useSendChatMessage() {
       console.error('No active thread')
       return
     }
-    setIsGeneratingResponse(true)
 
     if (engineParamsUpdate) setReloadModel(true)
 
@@ -361,7 +359,7 @@ export default function useSendChatMessage() {
       await waitForModelStarting(modelId)
       setQueuedMessage(false)
     }
-
+    setIsGeneratingResponse(true)
     events.emit(MessageEvent.OnMessageSent, messageRequest)
 
     setReloadModel(false)

web/hooks/useSettings.ts

@@ -70,11 +70,6 @@ export const useSettings = () => {
       }
     }
     await fs.writeFileSync(settingsFile, JSON.stringify(settings))
-
-    // Relaunch to apply settings
-    if (vulkan != null) {
-      window.location.reload()
-    }
   }
 
   return {

web/screens/Settings/Advanced/index.tsx

@@ -90,12 +90,38 @@ const Advanced = () => {
     [setPartialProxy, setProxy]
   )
 
-  const updateQuickAskEnabled = async (e: boolean) => {
+  const updateQuickAskEnabled = async (
+    e: boolean,
+    relaunch: boolean = true
+  ) => {
     const appConfiguration: AppConfiguration =
       await window.core?.api?.getAppConfigurations()
     appConfiguration.quick_ask = e
     await window.core?.api?.updateAppConfiguration(appConfiguration)
-    window.core?.api?.relaunch()
+    if (relaunch) window.core?.api?.relaunch()
+  }
+
+  const updateVulkanEnabled = async (e: boolean, relaunch: boolean = true) => {
+    toaster({
+      title: 'Reload',
+      description: 'Vulkan settings updated. Reload now to apply the changes.',
+    })
+    stopModel()
+    setVulkanEnabled(e)
+    await saveSettings({ vulkan: e, gpusInUse: [] })
+    // Relaunch to apply settings
+    if (relaunch) window.location.reload()
+  }
+
+  const updateExperimentalEnabled = async (e: boolean) => {
+    setExperimentalEnabled(e)
+    if (e) return
+
+    // It affects other settings, so we need to reset them
+    const isRelaunch = quickAskEnabled || vulkanEnabled
+    if (quickAskEnabled) await updateQuickAskEnabled(false, false)
+    if (vulkanEnabled) await updateVulkanEnabled(false, false)
+    if (isRelaunch) window.core?.api?.relaunch()
   }
 
   useEffect(() => {
@@ -179,7 +205,7 @@ const Advanced = () => {
           </div>
           <Switch
             checked={experimentalEnabled}
-            onCheckedChange={setExperimentalEnabled}
+            onCheckedChange={updateExperimentalEnabled}
           />
         </div>
 
@@ -381,16 +407,7 @@ const Advanced = () => {
 
             <Switch
               checked={vulkanEnabled}
-              onCheckedChange={(e) => {
+              onCheckedChange={(e) => updateVulkanEnabled(e)}
-                toaster({
-                  title: 'Reload',
-                  description:
-                    'Vulkan settings updated. Reload now to apply the changes.',
-                })
-                stopModel()
-                saveSettings({ vulkan: e, gpusInUse: [] })
-                setVulkanEnabled(e)
-              }}
             />
           </div>
         )}

web/screens/Settings/CoreExtensions/TensorRtExtensionItem.tsx

@@ -23,6 +23,8 @@ import { useAtomValue } from 'jotai'
 
 import { Marked, Renderer } from 'marked'
 
+import UpdateExtensionModal from './UpdateExtensionModal'
+
 import { extensionManager } from '@/extension'
 import Extension from '@/extension/Extension'
 import { installingExtensionAtom } from '@/helpers/atoms/Extension.atom'
@@ -39,7 +41,7 @@ const TensorRtExtensionItem: React.FC<Props> = ({ item }) => {
     useState<InstallationState>('NotRequired')
   const installingExtensions = useAtomValue(installingExtensionAtom)
   const [isGpuSupported, setIsGpuSupported] = useState<boolean>(false)
-
+  const [promptUpdateModal, setPromptUpdateModal] = useState<boolean>(false)
   const isInstalling = installingExtensions.some(
     (e) => e.extensionId === item.name
   )
@@ -69,7 +71,7 @@ const TensorRtExtensionItem: React.FC<Props> = ({ item }) => {
         return
       }
 
-      const supportedGpuArch = ['turing', 'ampere', 'ada']
+      const supportedGpuArch = ['ampere', 'ada']
       setIsGpuSupported(supportedGpuArch.includes(arch))
     }
     getSystemInfos()
@@ -138,6 +140,7 @@ const TensorRtExtensionItem: React.FC<Props> = ({ item }) => {
             installProgress={progress}
             installState={installState}
             onInstallClick={onInstallClick}
+            onUpdateClick={() => setPromptUpdateModal(true)}
             onCancelClick={onCancelInstallingClick}
           />
         </div>
@@ -177,6 +180,9 @@ const TensorRtExtensionItem: React.FC<Props> = ({ item }) => {
           </div>
         </div>
       )}
+      {promptUpdateModal && (
+        <UpdateExtensionModal onUpdateClick={onInstallClick} />
+      )}
     </div>
   )
 }
@@ -185,6 +191,7 @@ type InstallStateProps = {
   installProgress: number
   installState: InstallationState
   onInstallClick: () => void
+  onUpdateClick: () => void
   onCancelClick: () => void
 }
 
@@ -192,6 +199,7 @@ const InstallStateIndicator: React.FC<InstallStateProps> = ({
   installProgress,
   installState,
   onInstallClick,
+  onUpdateClick,
   onCancelClick,
 }) => {
   if (installProgress !== -1) {
@@ -218,6 +226,12 @@ const InstallStateIndicator: React.FC<InstallStateProps> = ({
           Installed
         </div>
       )
+    case 'Updatable':
+      return (
+        <Button themes="secondaryBlue" size="sm" onClick={onUpdateClick}>
+          Update
+        </Button>
+      )
     case 'NotInstalled':
       return (
         <Button themes="secondaryBlue" size="sm" onClick={onInstallClick}>

web/screens/Settings/CoreExtensions/UpdateExtensionModal.tsx

@@ -0,0 +1,58 @@
+import React from 'react'
+
+import {
+  Button,
+  Modal,
+  ModalClose,
+  ModalContent,
+  ModalFooter,
+  ModalHeader,
+  ModalPortal,
+  ModalTitle,
+  ModalTrigger,
+} from '@janhq/uikit'
+import { Paintbrush } from 'lucide-react'
+
+type Props = {
+  onUpdateClick: () => void
+}
+
+const UpdateExtensionModal: React.FC<Props> = ({ onUpdateClick }) => {
+  return (
+    <Modal>
+      <ModalTrigger asChild onClick={(e) => e.stopPropagation()}>
+        <div className="flex cursor-pointer items-center space-x-2 px-4 py-2 hover:bg-secondary">
+          <Paintbrush size={16} className="text-muted-foreground" />
+          <span className="text-bold text-black dark:text-muted-foreground">
+            Update extension
+          </span>
+        </div>
+      </ModalTrigger>
+      <ModalPortal />
+      <ModalContent>
+        <ModalHeader>
+          <ModalTitle>Clean Thread</ModalTitle>
+        </ModalHeader>
+        <p>
+          Updating this extension may result in the loss of any custom models or
+          data associated with the current version. We recommend backing up any
+          important data before proceeding with the update.
+        </p>
+        <ModalFooter>
+          <div className="flex gap-x-2">
+            <ModalClose asChild onClick={(e) => e.stopPropagation()}>
+              <Button themes="ghost">No</Button>
+            </ModalClose>
+            <ModalClose asChild>
+              <Button themes="danger" onClick={onUpdateClick} autoFocus>
+                Yes
+              </Button>
+            </ModalClose>
+          </div>
+        </ModalFooter>
+      </ModalContent>
+    </Modal>
+  )
+}
+
+export default React.memo(UpdateExtensionModal)