diff --git a/packages/shared/utils.ts b/packages/shared/utils.ts
index e87e2f2bef..a14f78958d 100644
--- a/packages/shared/utils.ts
+++ b/packages/shared/utils.ts
@@ -4,3 +4,34 @@ export const defaultAppHeaders = () => {
     'X-Title': 'Cherry Studio'
   }
 }
+
+// Following two function are not being used for now.
+// I may use them in the future, so just keep them commented. - by eurfelux
+
+/**
+ * Converts an `undefined` value to `null`, otherwise returns the value as-is.
+ * @param value - The value to check
+ * @returns `null` if the input is `undefined`; otherwise the input value
+ */
+
+// export function toNullIfUndefined<T>(value: T | undefined): T | null {
+//   if (value === undefined) {
+//     return null
+//   } else {
+//     return value
+//   }
+// }
+
+/**
+ * Converts a `null` value to `undefined`, otherwise returns the value as-is.
+ * @param value - The value to check
+ * @returns `undefined` if the input is `null`; otherwise the input value
+ */
+
+// export function toUndefinedIfNull<T>(value: T | null): T | undefined {
+//   if (value === null) {
+//     return undefined
+//   } else {
+//     return value
+//   }
+// }
diff --git a/src/renderer/src/aiCore/legacy/clients/BaseApiClient.ts b/src/renderer/src/aiCore/legacy/clients/BaseApiClient.ts
index f520162496..1caf483205 100644
--- a/src/renderer/src/aiCore/legacy/clients/BaseApiClient.ts
+++ b/src/renderer/src/aiCore/legacy/clients/BaseApiClient.ts
@@ -19,7 +19,6 @@ import type {
   MCPToolResponse,
   MemoryItem,
   Model,
-  OpenAIVerbosity,
   Provider,
   ToolCallResponse,
   WebSearchProviderResponse,
@@ -33,6 +32,7 @@ import {
   OpenAIServiceTiers,
   SystemProviderIds
 } from '@renderer/types'
+import type { OpenAIVerbosity } from '@renderer/types/aiCoreTypes'
 import type { Message } from '@renderer/types/newMessage'
 import type {
   RequestOptions,
diff --git a/src/renderer/src/aiCore/prepareParams/parameterBuilder.ts b/src/renderer/src/aiCore/prepareParams/parameterBuilder.ts
index d55dd9d55e..6f8747a7c5 100644
--- a/src/renderer/src/aiCore/prepareParams/parameterBuilder.ts
+++ b/src/renderer/src/aiCore/prepareParams/parameterBuilder.ts
@@ -61,7 +61,7 @@ export async function buildStreamTextParams(
       timeout?: number
       headers?: Record<string, string>
     }
-  } = {}
+  }
 ): Promise<{
   params: StreamTextParams
   modelId: string
diff --git a/src/renderer/src/aiCore/utils/options.ts b/src/renderer/src/aiCore/utils/options.ts
index 7f4cd33608..2dc142cc46 100644
--- a/src/renderer/src/aiCore/utils/options.ts
+++ b/src/renderer/src/aiCore/utils/options.ts
@@ -1,3 +1,7 @@
+import type { AnthropicProviderOptions } from '@ai-sdk/anthropic'
+import type { GoogleGenerativeAIProviderOptions } from '@ai-sdk/google'
+import type { OpenAIResponsesProviderOptions } from '@ai-sdk/openai'
+import type { XaiProviderOptions } from '@ai-sdk/xai'
 import { baseProviderIdSchema, customProviderIdSchema } from '@cherrystudio/ai-core/provider'
 import { loggerService } from '@logger'
 import {
@@ -9,15 +13,28 @@ import {
 } from '@renderer/config/models'
 import { isSupportServiceTierProvider } from '@renderer/config/providers'
 import { mapLanguageToQwenMTModel } from '@renderer/config/translate'
-import type { Assistant, Model, Provider } from '@renderer/types'
+import { getStoreSetting } from '@renderer/hooks/useSettings'
+import type { RootState } from '@renderer/store'
+import type {
+  Assistant,
+  GroqServiceTier,
+  GroqSystemProvider,
+  Model,
+  NotGroqProvider,
+  OpenAIServiceTier,
+  Provider,
+  ServiceTier
+} from '@renderer/types'
 import {
   GroqServiceTiers,
   isGroqServiceTier,
+  isGroqSystemProvider,
   isOpenAIServiceTier,
   isTranslateAssistant,
-  OpenAIServiceTiers,
-  SystemProviderIds
+  OpenAIServiceTiers
 } from '@renderer/types'
+import type { OpenAIVerbosity } from '@renderer/types/aiCoreTypes'
+import type { JSONValue } from 'ai'
 import { t } from 'i18next'
 
 import { getAiSdkProviderId } from '../provider/factory'
@@ -35,8 +52,31 @@ import { getWebSearchParams } from './websearch'
 
 const logger = loggerService.withContext('aiCore.utils.options')
 
-// copy from BaseApiClient.ts
-const getServiceTier = (model: Model, provider: Provider) => {
+function toOpenAIServiceTier(model: Model, serviceTier: ServiceTier): OpenAIServiceTier {
+  if (
+    !isOpenAIServiceTier(serviceTier) ||
+    (serviceTier === OpenAIServiceTiers.flex && !isSupportFlexServiceTierModel(model))
+  ) {
+    return undefined
+  } else {
+    return serviceTier
+  }
+}
+
+function toGroqServiceTier(model: Model, serviceTier: ServiceTier): GroqServiceTier {
+  if (
+    !isGroqServiceTier(serviceTier) ||
+    (serviceTier === GroqServiceTiers.flex && !isSupportFlexServiceTierModel(model))
+  ) {
+    return undefined
+  } else {
+    return serviceTier
+  }
+}
+
+function getServiceTier<T extends GroqSystemProvider>(model: Model, provider: T): GroqServiceTier
+function getServiceTier<T extends NotGroqProvider>(model: Model, provider: T): OpenAIServiceTier
+function getServiceTier<T extends Provider>(model: Model, provider: T): OpenAIServiceTier | GroqServiceTier {
   const serviceTierSetting = provider.serviceTier
 
   if (!isSupportServiceTierProvider(provider) || !isOpenAIModel(model) || !serviceTierSetting) {
@@ -44,24 +84,17 @@ const getServiceTier = (model: Model, provider: Provider) => {
   }
 
   // 处理不同供应商需要 fallback 到默认值的情况
-  if (provider.id === SystemProviderIds.groq) {
-    if (
-      !isGroqServiceTier(serviceTierSetting) ||
-      (serviceTierSetting === GroqServiceTiers.flex && !isSupportFlexServiceTierModel(model))
-    ) {
-      return undefined
-    }
+  if (isGroqSystemProvider(provider)) {
+    return toGroqServiceTier(model, serviceTierSetting)
   } else {
     // 其他 OpenAI 供应商，假设他们的服务层级设置和 OpenAI 完全相同
-    if (
-      !isOpenAIServiceTier(serviceTierSetting) ||
-      (serviceTierSetting === OpenAIServiceTiers.flex && !isSupportFlexServiceTierModel(model))
-    ) {
-      return undefined
-    }
+    return toOpenAIServiceTier(model, serviceTierSetting)
   }
+}
 
-  return serviceTierSetting
+function getVerbosity(): OpenAIVerbosity {
+  const openAI = getStoreSetting('openAI')
+  return openAI.verbosity
 }
 
 /**
@@ -78,13 +111,13 @@ export function buildProviderOptions(
     enableWebSearch: boolean
     enableGenerateImage: boolean
   }
-): Record<string, any> {
+): Record<string, Record<string, JSONValue>> {
   logger.debug('buildProviderOptions', { assistant, model, actualProvider, capabilities })
   const rawProviderId = getAiSdkProviderId(actualProvider)
   // 构建 provider 特定的选项
   let providerSpecificOptions: Record<string, any> = {}
-  const serviceTierSetting = getServiceTier(model, actualProvider)
-  providerSpecificOptions.serviceTier = serviceTierSetting
+  const serviceTier = getServiceTier(model, actualProvider)
+  const textVerbosity = getVerbosity()
   // 根据 provider 类型分离构建逻辑
   const { data: baseProviderId, success } = baseProviderIdSchema.safeParse(rawProviderId)
   if (success) {
@@ -94,9 +127,14 @@ export function buildProviderOptions(
       case 'openai-chat':
       case 'azure':
       case 'azure-responses':
-        providerSpecificOptions = {
-          ...buildOpenAIProviderOptions(assistant, model, capabilities),
-          serviceTier: serviceTierSetting
+        {
+          const options: OpenAIResponsesProviderOptions = buildOpenAIProviderOptions(
+            assistant,
+            model,
+            capabilities,
+            serviceTier
+          )
+          providerSpecificOptions = options
         }
         break
       case 'anthropic':
@@ -116,12 +154,19 @@ export function buildProviderOptions(
         // 对于其他 provider，使用通用的构建逻辑
         providerSpecificOptions = {
           ...buildGenericProviderOptions(assistant, model, capabilities),
-          serviceTier: serviceTierSetting
+          serviceTier,
+          textVerbosity
         }
         break
       }
       case 'cherryin':
-        providerSpecificOptions = buildCherryInProviderOptions(assistant, model, capabilities, actualProvider)
+        providerSpecificOptions = buildCherryInProviderOptions(
+          assistant,
+          model,
+          capabilities,
+          actualProvider,
+          serviceTier
+        )
         break
       default:
         throw new Error(`Unsupported base provider ${baseProviderId}`)
@@ -142,13 +187,14 @@ export function buildProviderOptions(
           providerSpecificOptions = buildBedrockProviderOptions(assistant, model, capabilities)
           break
         case 'huggingface':
-          providerSpecificOptions = buildOpenAIProviderOptions(assistant, model, capabilities)
+          providerSpecificOptions = buildOpenAIProviderOptions(assistant, model, capabilities, serviceTier)
           break
         default:
           // 对于其他 provider，使用通用的构建逻辑
           providerSpecificOptions = {
             ...buildGenericProviderOptions(assistant, model, capabilities),
-            serviceTier: serviceTierSetting
+            serviceTier,
+            textVerbosity
           }
       }
     } else {
@@ -189,10 +235,12 @@ function buildOpenAIProviderOptions(
     enableReasoning: boolean
     enableWebSearch: boolean
     enableGenerateImage: boolean
-  }
-): Record<string, any> {
+  },
+  serviceTier: OpenAIServiceTier
+): OpenAIResponsesProviderOptions {
   const { enableReasoning } = capabilities
   let providerOptions: Record<string, any> = {}
+
   // OpenAI 推理参数
   if (enableReasoning) {
     const reasoningParams = getOpenAIReasoningParams(assistant, model)
@@ -203,7 +251,7 @@ function buildOpenAIProviderOptions(
   }
 
   if (isSupportVerbosityModel(model)) {
-    const state = window.store?.getState()
+    const state: RootState = window.store?.getState()
     const userVerbosity = state?.settings?.openAI?.verbosity
 
     if (userVerbosity && ['low', 'medium', 'high'].includes(userVerbosity)) {
@@ -218,6 +266,11 @@ function buildOpenAIProviderOptions(
     }
   }
 
+  providerOptions = {
+    ...providerOptions,
+    serviceTier
+  }
+
   return providerOptions
 }
 
@@ -232,7 +285,7 @@ function buildAnthropicProviderOptions(
     enableWebSearch: boolean
     enableGenerateImage: boolean
   }
-): Record<string, any> {
+): AnthropicProviderOptions {
   const { enableReasoning } = capabilities
   let providerOptions: Record<string, any> = {}
 
@@ -259,7 +312,7 @@ function buildGeminiProviderOptions(
     enableWebSearch: boolean
     enableGenerateImage: boolean
   }
-): Record<string, any> {
+): GoogleGenerativeAIProviderOptions {
   const { enableReasoning, enableGenerateImage } = capabilities
   let providerOptions: Record<string, any> = {}
 
@@ -290,7 +343,7 @@ function buildXAIProviderOptions(
     enableWebSearch: boolean
     enableGenerateImage: boolean
   }
-): Record<string, any> {
+): XaiProviderOptions {
   const { enableReasoning } = capabilities
   let providerOptions: Record<string, any> = {}
 
@@ -313,16 +366,12 @@ function buildCherryInProviderOptions(
     enableWebSearch: boolean
     enableGenerateImage: boolean
   },
-  actualProvider: Provider
-): Record<string, any> {
-  const serviceTierSetting = getServiceTier(model, actualProvider)
-
+  actualProvider: Provider,
+  serviceTier: OpenAIServiceTier
+): OpenAIResponsesProviderOptions | AnthropicProviderOptions | GoogleGenerativeAIProviderOptions {
   switch (actualProvider.type) {
     case 'openai':
-      return {
-        ...buildOpenAIProviderOptions(assistant, model, capabilities),
-        serviceTier: serviceTierSetting
-      }
+      return buildOpenAIProviderOptions(assistant, model, capabilities, serviceTier)
 
     case 'anthropic':
       return buildAnthropicProviderOptions(assistant, model, capabilities)
diff --git a/src/renderer/src/aiCore/utils/reasoning.ts b/src/renderer/src/aiCore/utils/reasoning.ts
index dfe084179c..f261f71a7a 100644
--- a/src/renderer/src/aiCore/utils/reasoning.ts
+++ b/src/renderer/src/aiCore/utils/reasoning.ts
@@ -1,6 +1,7 @@
 import type { BedrockProviderOptions } from '@ai-sdk/amazon-bedrock'
 import type { AnthropicProviderOptions } from '@ai-sdk/anthropic'
 import type { GoogleGenerativeAIProviderOptions } from '@ai-sdk/google'
+import type { OpenAIResponsesProviderOptions } from '@ai-sdk/openai'
 import type { XaiProviderOptions } from '@ai-sdk/xai'
 import { loggerService } from '@logger'
 import { DEFAULT_MAX_TOKENS } from '@renderer/config/constant'
@@ -35,9 +36,9 @@ import {
 import { isSupportEnableThinkingProvider } from '@renderer/config/providers'
 import { getStoreSetting } from '@renderer/hooks/useSettings'
 import { getAssistantSettings, getProviderByModel } from '@renderer/services/AssistantService'
-import type { SettingsState } from '@renderer/store/settings'
 import type { Assistant, Model } from '@renderer/types'
 import { EFFORT_RATIO, isSystemProvider, SystemProviderIds } from '@renderer/types'
+import type { OpenAISummaryText } from '@renderer/types/aiCoreTypes'
 import type { ReasoningEffortOptionalParams } from '@renderer/types/sdk'
 import { toInteger } from 'lodash'
 
@@ -341,10 +342,14 @@ export function getReasoningEffort(assistant: Assistant, model: Model): Reasonin
 }
 
 /**
- * 获取 OpenAI 推理参数
- * 从 OpenAIResponseAPIClient 和 OpenAIAPIClient 中提取的逻辑
+ * Get OpenAI reasoning parameters
+ * Extracted from OpenAIResponseAPIClient and OpenAIAPIClient logic
+ * For official OpenAI provider only
  */
-export function getOpenAIReasoningParams(assistant: Assistant, model: Model): Record<string, any> {
+export function getOpenAIReasoningParams(
+  assistant: Assistant,
+  model: Model
+): Pick<OpenAIResponsesProviderOptions, 'reasoningEffort' | 'reasoningSummary'> {
   if (!isReasoningModel(model)) {
     return {}
   }
@@ -355,6 +360,10 @@ export function getOpenAIReasoningParams(assistant: Assistant, model: Model): Re
     return {}
   }
 
+  if (isOpenAIDeepResearchModel(model) || reasoningEffort === 'auto') {
+    reasoningEffort = 'medium'
+  }
+
   // 非OpenAI模型，但是Provider类型是responses/azure openai的情况
   if (!isOpenAIModel(model)) {
     return {
@@ -362,21 +371,17 @@ export function getOpenAIReasoningParams(assistant: Assistant, model: Model): Re
     }
   }
 
-  const openAI = getStoreSetting('openAI') as SettingsState['openAI']
-  const summaryText = openAI?.summaryText || 'off'
+  const openAI = getStoreSetting('openAI')
+  const summaryText = openAI.summaryText
 
-  let reasoningSummary: string | undefined = undefined
+  let reasoningSummary: OpenAISummaryText = undefined
 
-  if (summaryText === 'off' || model.id.includes('o1-pro')) {
+  if (model.id.includes('o1-pro')) {
     reasoningSummary = undefined
   } else {
     reasoningSummary = summaryText
   }
 
-  if (isOpenAIDeepResearchModel(model)) {
-    reasoningEffort = 'medium'
-  }
-
   // OpenAI 推理参数
   if (isSupportedReasoningEffortOpenAIModel(model)) {
     return {
diff --git a/src/renderer/src/components/Selector.tsx b/src/renderer/src/components/Selector.tsx
index e30bc64193..38567fc200 100644
--- a/src/renderer/src/components/Selector.tsx
+++ b/src/renderer/src/components/Selector.tsx
@@ -6,7 +6,7 @@ import { useEffect, useMemo, useRef, useState } from 'react'
 import { useTranslation } from 'react-i18next'
 import styled, { css } from 'styled-components'
 
-interface SelectorOption<V = string | number> {
+interface SelectorOption<V = string | number | undefined | null> {
   label: string | ReactNode
   value: V
   type?: 'group'
@@ -14,7 +14,7 @@ interface SelectorOption<V = string | number> {
   disabled?: boolean
 }
 
-interface BaseSelectorProps<V = string | number> {
+interface BaseSelectorProps<V = string | number | undefined | null> {
   options: SelectorOption<V>[]
   placeholder?: string
   placement?: 'topLeft' | 'topCenter' | 'topRight' | 'bottomLeft' | 'bottomCenter' | 'bottomRight' | 'top' | 'bottom'
@@ -39,7 +39,7 @@ interface MultipleSelectorProps<V> extends BaseSelectorProps<V> {
 
 export type SelectorProps<V> = SingleSelectorProps<V> | MultipleSelectorProps<V>
 
-const Selector = <V extends string | number>({
+const Selector = <V extends string | number | undefined | null>({
   options,
   value,
   onChange = () => {},
diff --git a/src/renderer/src/config/models/utils.ts b/src/renderer/src/config/models/utils.ts
index 7fb7c61362..6c75d49251 100644
--- a/src/renderer/src/config/models/utils.ts
+++ b/src/renderer/src/config/models/utils.ts
@@ -1,6 +1,7 @@
 import type OpenAI from '@cherrystudio/openai'
 import { isEmbeddingModel, isRerankModel } from '@renderer/config/models/embedding'
 import type { Model } from '@renderer/types'
+import type { OpenAIVerbosity, ValidOpenAIVerbosity } from '@renderer/types/aiCoreTypes'
 import { getLowerBaseModelName } from '@renderer/utils'
 
 import { WEB_SEARCH_PROMPT_FOR_OPENROUTER } from '../prompts'
@@ -242,17 +243,20 @@ export const isGPT51SeriesModel = (model: Model) => {
 
 // GPT-5 verbosity configuration
 // gpt-5-pro only supports 'high', other GPT-5 models support all levels
-export const MODEL_SUPPORTED_VERBOSITY: Record<string, ('low' | 'medium' | 'high')[]> = {
+export const MODEL_SUPPORTED_VERBOSITY: Record<string, ValidOpenAIVerbosity[]> = {
   'gpt-5-pro': ['high'],
   default: ['low', 'medium', 'high']
-}
+} as const
 
-export const getModelSupportedVerbosity = (model: Model): ('low' | 'medium' | 'high')[] => {
+export const getModelSupportedVerbosity = (model: Model): OpenAIVerbosity[] => {
   const modelId = getLowerBaseModelName(model.id)
+  let supportedValues: ValidOpenAIVerbosity[]
   if (modelId.includes('gpt-5-pro')) {
-    return MODEL_SUPPORTED_VERBOSITY['gpt-5-pro']
+    supportedValues = MODEL_SUPPORTED_VERBOSITY['gpt-5-pro']
+  } else {
+    supportedValues = MODEL_SUPPORTED_VERBOSITY.default
   }
-  return MODEL_SUPPORTED_VERBOSITY.default
+  return [undefined, ...supportedValues]
 }
 
 export const isGeminiModel = (model: Model) => {
diff --git a/src/renderer/src/i18n/locales/en-us.json b/src/renderer/src/i18n/locales/en-us.json
index ab3c8aa9a1..5b1de2a257 100644
--- a/src/renderer/src/i18n/locales/en-us.json
+++ b/src/renderer/src/i18n/locales/en-us.json
@@ -1158,6 +1158,7 @@
     "name": "Name",
     "no_results": "No results",
     "none": "None",
+    "off": "Off",
     "open": "Open",
     "paste": "Paste",
     "placeholders": {
@@ -4259,7 +4260,6 @@
         "default": "default",
         "flex": "flex",
         "on_demand": "on demand",
-        "performance": "performance",
         "priority": "priority",
         "tip": "Specifies the latency tier to use for processing the request",
         "title": "Service Tier"
@@ -4278,7 +4278,7 @@
         "low": "Low",
         "medium": "Medium",
         "tip": "Control the level of detail in the model's output",
-        "title": "Level of detail"
+        "title": "Verbosity"
       }
     },
     "privacy": {
diff --git a/src/renderer/src/i18n/locales/zh-cn.json b/src/renderer/src/i18n/locales/zh-cn.json
index ff724e1cc8..8d7073fcfd 100644
--- a/src/renderer/src/i18n/locales/zh-cn.json
+++ b/src/renderer/src/i18n/locales/zh-cn.json
@@ -1158,6 +1158,7 @@
     "name": "名称",
     "no_results": "无结果",
     "none": "无",
+    "off": "关闭",
     "open": "打开",
     "paste": "粘贴",
     "placeholders": {
@@ -4259,7 +4260,6 @@
         "default": "默认",
         "flex": "灵活",
         "on_demand": "按需",
-        "performance": "性能",
         "priority": "优先",
         "tip": "指定用于处理请求的延迟层级",
         "title": "服务层级"
diff --git a/src/renderer/src/i18n/locales/zh-tw.json b/src/renderer/src/i18n/locales/zh-tw.json
index 76e916a9a7..72eb71ea97 100644
--- a/src/renderer/src/i18n/locales/zh-tw.json
+++ b/src/renderer/src/i18n/locales/zh-tw.json
@@ -1158,6 +1158,7 @@
     "name": "名稱",
     "no_results": "沒有結果",
     "none": "無",
+    "off": "關閉",
     "open": "開啟",
     "paste": "貼上",
     "placeholders": {
@@ -4259,7 +4260,6 @@
         "default": "預設",
         "flex": "彈性",
         "on_demand": "按需",
-        "performance": "效能",
         "priority": "優先",
         "tip": "指定用於處理請求的延遲層級",
         "title": "服務層級"
diff --git a/src/renderer/src/i18n/translate/pt-pt.json b/src/renderer/src/i18n/translate/pt-pt.json
index 541a728946..1ca373f394 100644
--- a/src/renderer/src/i18n/translate/pt-pt.json
+++ b/src/renderer/src/i18n/translate/pt-pt.json
@@ -1158,6 +1158,7 @@
     "name": "Nome",
     "no_results": "Nenhum resultado",
     "none": "Nenhum",
+    "off": "Desligado",
     "open": "Abrir",
     "paste": "Colar",
     "placeholders": {
@@ -4223,7 +4224,6 @@
         "default": "Padrão",
         "flex": "Flexível",
         "on_demand": "sob demanda",
-        "performance": "desempenho",
         "priority": "prioridade",
         "tip": "Especifique o nível de latência usado para processar a solicitação",
         "title": "Nível de Serviço"
diff --git a/src/renderer/src/pages/home/Tabs/components/OpenAISettingsGroup.tsx b/src/renderer/src/pages/home/Tabs/components/OpenAISettingsGroup.tsx
index 2960724183..b6ecf88c72 100644
--- a/src/renderer/src/pages/home/Tabs/components/OpenAISettingsGroup.tsx
+++ b/src/renderer/src/pages/home/Tabs/components/OpenAISettingsGroup.tsx
@@ -12,9 +12,9 @@ import { CollapsibleSettingGroup } from '@renderer/pages/settings/SettingGroup'
 import type { RootState } from '@renderer/store'
 import { useAppDispatch } from '@renderer/store'
 import { setOpenAISummaryText, setOpenAIVerbosity } from '@renderer/store/settings'
-import type { Model, OpenAIServiceTier, OpenAISummaryText, ServiceTier } from '@renderer/types'
+import type { GroqServiceTier, Model, OpenAIServiceTier, ServiceTier } from '@renderer/types'
 import { GroqServiceTiers, OpenAIServiceTiers, SystemProviderIds } from '@renderer/types'
-import type { OpenAIVerbosity } from '@types'
+import type { OpenAISummaryText, OpenAIVerbosity } from '@renderer/types/aiCoreTypes'
 import { Tooltip } from 'antd'
 import { CircleHelp } from 'lucide-react'
 import type { FC } from 'react'
@@ -22,6 +22,21 @@ import { useCallback, useEffect, useMemo } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useSelector } from 'react-redux'
 
+type VerbosityOption = {
+  value: OpenAIVerbosity
+  label: string
+}
+
+type SummaryTextOption = {
+  value: OpenAISummaryText
+  label: string
+}
+
+type OpenAIServiceTierOption = { value: OpenAIServiceTier; label: string }
+type GroqServiceTierOption = { value: GroqServiceTier; label: string }
+
+type ServiceTierOptions = OpenAIServiceTierOption[] | GroqServiceTierOption[]
+
 interface Props {
   model: Model
   providerId: string
@@ -67,6 +82,10 @@ const OpenAISettingsGroup: FC<Props> = ({ model, providerId, SettingGroup, Setti
   )
 
   const summaryTextOptions = [
+    {
+      value: undefined,
+      label: t('common.default')
+    },
     {
       value: 'auto',
       label: t('settings.openai.summary_text_mode.auto')
@@ -76,13 +95,17 @@ const OpenAISettingsGroup: FC<Props> = ({ model, providerId, SettingGroup, Setti
       label: t('settings.openai.summary_text_mode.detailed')
     },
     {
-      value: 'off',
-      label: t('settings.openai.summary_text_mode.off')
+      value: 'concise',
+      label: t('settings.openai.summary_text_mode.concise')
     }
-  ]
+  ] as const satisfies SummaryTextOption[]
 
   const verbosityOptions = useMemo(() => {
     const allOptions = [
+      {
+        value: undefined,
+        label: t('common.default')
+      },
       {
         value: 'low',
         label: t('settings.openai.verbosity.low')
@@ -95,15 +118,23 @@ const OpenAISettingsGroup: FC<Props> = ({ model, providerId, SettingGroup, Setti
         value: 'high',
         label: t('settings.openai.verbosity.high')
       }
-    ]
+    ] as const satisfies VerbosityOption[]
     const supportedVerbosityLevels = getModelSupportedVerbosity(model)
-    return allOptions.filter((option) => supportedVerbosityLevels.includes(option.value as any))
+    return allOptions.filter((option) => supportedVerbosityLevels.includes(option.value))
   }, [model, t])
 
   const serviceTierOptions = useMemo(() => {
-    let baseOptions: { value: ServiceTier; label: string }[]
+    let options: ServiceTierOptions
     if (provider.id === SystemProviderIds.groq) {
-      baseOptions = [
+      options = [
+        {
+          value: null,
+          label: t('common.off')
+        },
+        {
+          value: undefined,
+          label: t('common.default')
+        },
         {
           value: 'auto',
           label: t('settings.openai.service_tier.auto')
@@ -115,15 +146,11 @@ const OpenAISettingsGroup: FC<Props> = ({ model, providerId, SettingGroup, Setti
         {
           value: 'flex',
           label: t('settings.openai.service_tier.flex')
-        },
-        {
-          value: 'performance',
-          label: t('settings.openai.service_tier.performance')
         }
-      ]
+      ] as const satisfies GroqServiceTierOption[]
     } else {
       // 其他情况默认是和 OpenAI 相同
-      baseOptions = [
+      options = [
         {
           value: 'auto',
           label: t('settings.openai.service_tier.auto')
@@ -140,9 +167,9 @@ const OpenAISettingsGroup: FC<Props> = ({ model, providerId, SettingGroup, Setti
           value: 'priority',
           label: t('settings.openai.service_tier.priority')
         }
-      ]
+      ] as const satisfies OpenAIServiceTierOption[]
     }
-    return baseOptions.filter((option) => {
+    return options.filter((option) => {
       if (option.value === 'flex') {
         return isSupportedFlexServiceTier
       }
diff --git a/src/renderer/src/services/ApiService.ts b/src/renderer/src/services/ApiService.ts
index 10e191bb38..f19c90b61f 100644
--- a/src/renderer/src/services/ApiService.ts
+++ b/src/renderer/src/services/ApiService.ts
@@ -83,7 +83,7 @@ export async function fetchChatCompletion({
   messages,
   prompt,
   assistant,
-  options,
+  requestOptions,
   onChunkReceived,
   topicId,
   uiMessages
@@ -124,7 +124,7 @@ export async function fetchChatCompletion({
   } = await buildStreamTextParams(messages, assistant, provider, {
     mcpTools: mcpTools,
     webSearchProviderId: assistant.webSearchProviderId,
-    requestOptions: options
+    requestOptions
   })
 
   // Safely fallback to prompt tool use when function calling is not supported by model.
diff --git a/src/renderer/src/services/OrchestrateService.ts b/src/renderer/src/services/OrchestrateService.ts
index 1f365b39b6..71f17d6804 100644
--- a/src/renderer/src/services/OrchestrateService.ts
+++ b/src/renderer/src/services/OrchestrateService.ts
@@ -48,7 +48,7 @@ export class OrchestrationService {
       await fetchChatCompletion({
         messages: modelMessages,
         assistant: assistant,
-        options: request.options,
+        requestOptions: request.options,
         onChunkReceived,
         topicId: request.topicId,
         uiMessages: uiMessages
@@ -80,7 +80,7 @@ export async function transformMessagesAndFetch(
     await fetchChatCompletion({
       messages: modelMessages,
       assistant: assistant,
-      options: request.options,
+      requestOptions: request.options,
       onChunkReceived,
       topicId: request.topicId,
       uiMessages
diff --git a/src/renderer/src/services/TranslateService.ts b/src/renderer/src/services/TranslateService.ts
index f7abfdb3b9..a5abb2baee 100644
--- a/src/renderer/src/services/TranslateService.ts
+++ b/src/renderer/src/services/TranslateService.ts
@@ -2,7 +2,7 @@ import { loggerService } from '@logger'
 import { db } from '@renderer/databases'
 import type {
   CustomTranslateLanguage,
-  FetchChatCompletionOptions,
+  FetchChatCompletionRequestOptions,
   TranslateHistory,
   TranslateLanguage,
   TranslateLanguageCode
@@ -56,15 +56,15 @@ export const translateText = async (
     onResponse?.(translatedText, completed)
   }
 
-  const options = {
+  const requestOptions = {
     signal
-  } satisfies FetchChatCompletionOptions
+  } satisfies FetchChatCompletionRequestOptions
 
   try {
     await fetchChatCompletion({
       prompt: assistant.content,
       assistant,
-      options,
+      requestOptions,
       onChunkReceived: onChunk
     })
   } catch (e) {
diff --git a/src/renderer/src/store/index.ts b/src/renderer/src/store/index.ts
index 16254dfaa8..2bb9079370 100644
--- a/src/renderer/src/store/index.ts
+++ b/src/renderer/src/store/index.ts
@@ -67,7 +67,7 @@ const persistedReducer = persistReducer(
   {
     key: 'cherry-studio',
     storage,
-    version: 176,
+    version: 177,
     blacklist: ['runtime', 'messages', 'messageBlocks', 'tabs', 'toolPermissions'],
     migrate
   },
diff --git a/src/renderer/src/store/migrate.ts b/src/renderer/src/store/migrate.ts
index 8b8b00d20e..13755fdaf1 100644
--- a/src/renderer/src/store/migrate.ts
+++ b/src/renderer/src/store/migrate.ts
@@ -1499,6 +1499,7 @@ const migrateConfig = {
   '102': (state: RootState) => {
     try {
       state.settings.openAI = {
+        // @ts-expect-error it's a removed type. migrated on 177
         summaryText: 'off',
         serviceTier: 'auto',
         verbosity: 'medium'
@@ -1592,6 +1593,7 @@ const migrateConfig = {
       addMiniApp(state, 'google')
       if (!state.settings.openAI) {
         state.settings.openAI = {
+          // @ts-expect-error it's a removed type. migrated on 177
           summaryText: 'off',
           serviceTier: 'auto',
           verbosity: 'medium'
@@ -2856,6 +2858,19 @@ const migrateConfig = {
       logger.error('migrate 176 error', error as Error)
       return state
     }
+  },
+  '177': (state: RootState) => {
+    try {
+      // @ts-expect-error it's a removed type
+      if (state.settings.openAI.summaryText === 'off') {
+        state.settings.openAI.summaryText = 'auto'
+      }
+      logger.info('migrate 177 success')
+      return state
+    } catch (error) {
+      logger.error('migrate 177 error', error as Error)
+      return state
+    }
   }
 }
 
diff --git a/src/renderer/src/store/settings.ts b/src/renderer/src/store/settings.ts
index 45f521b3df..cb871d37e6 100644
--- a/src/renderer/src/store/settings.ts
+++ b/src/renderer/src/store/settings.ts
@@ -10,16 +10,15 @@ import type {
   LanguageVarious,
   MathEngine,
   OpenAIServiceTier,
-  OpenAISummaryText,
   PaintingProvider,
   S3Config,
   SidebarIcon,
   TranslateLanguageCode
 } from '@renderer/types'
 import { ThemeMode } from '@renderer/types'
+import type { OpenAISummaryText, OpenAIVerbosity } from '@renderer/types/aiCoreTypes'
 import { uuid } from '@renderer/utils'
 import { UpgradeChannel } from '@shared/config/constant'
-import type { OpenAIVerbosity } from '@types'
 
 import type { RemoteSyncState } from './backup'
 
@@ -375,7 +374,7 @@ export const initialState: SettingsState = {
   },
   // OpenAI
   openAI: {
-    summaryText: 'off',
+    summaryText: 'auto',
     serviceTier: 'auto',
     verbosity: 'medium'
   },
diff --git a/src/renderer/src/types/aiCoreTypes.ts b/src/renderer/src/types/aiCoreTypes.ts
index a2ff5a4cef..6327fe6835 100644
--- a/src/renderer/src/types/aiCoreTypes.ts
+++ b/src/renderer/src/types/aiCoreTypes.ts
@@ -1,3 +1,5 @@
+import type OpenAI from '@cherrystudio/openai'
+import type { NotNull, NotUndefined } from '@types'
 import type { ImageModel, LanguageModel } from 'ai'
 import type { generateObject, generateText, ModelMessage, streamObject, streamText } from 'ai'
 
@@ -27,3 +29,16 @@ export type StreamObjectParams = Omit<Parameters<typeof streamObject>[0], 'model
 export type GenerateObjectParams = Omit<Parameters<typeof generateObject>[0], 'model'>
 
 export type AiSdkModel = LanguageModel | ImageModel
+
+// The original type unite both undefined and null.
+// I pick undefined as the unique falsy type since they seem like share the same meaning according to OpenAI API docs.
+// Parameter would not be passed into request if it's undefined.
+export type OpenAIVerbosity = NotNull<OpenAI.Responses.ResponseTextConfig['verbosity']>
+export type ValidOpenAIVerbosity = NotUndefined<OpenAIVerbosity>
+
+export type OpenAIReasoningEffort = OpenAI.ReasoningEffort
+
+// The original type unite both undefined and null.
+// I pick undefined as the unique falsy type since they seem like share the same meaning according to OpenAI API docs.
+// Parameter would not be passed into request if it's undefined.
+export type OpenAISummaryText = NotNull<OpenAI.Reasoning['summary']>
diff --git a/src/renderer/src/types/index.ts b/src/renderer/src/types/index.ts
index 01d654fdb2..2ec88765fc 100644
--- a/src/renderer/src/types/index.ts
+++ b/src/renderer/src/types/index.ts
@@ -871,10 +871,6 @@ export interface StoreSyncAction {
   }
 }
 
-export type OpenAIVerbosity = 'high' | 'medium' | 'low'
-
-export type OpenAISummaryText = 'auto' | 'concise' | 'detailed' | 'off'
-
 export type S3Config = {
   endpoint: string
   region: string
@@ -1091,7 +1087,7 @@ export const isHexColor = (value: string): value is HexColor => {
   return /^#([0-9A-F]{3}){1,2}$/i.test(value)
 }
 
-export type FetchChatCompletionOptions = {
+export type FetchChatCompletionRequestOptions = {
   signal?: AbortSignal
   timeout?: number
   headers?: Record<string, string>
@@ -1099,7 +1095,7 @@ export type FetchChatCompletionOptions = {
 
 type BaseParams = {
   assistant: Assistant
-  options?: FetchChatCompletionOptions
+  requestOptions?: FetchChatCompletionRequestOptions
   onChunkReceived: (chunk: Chunk) => void
   topicId?: string // 添加 topicId 参数
   uiMessages?: Message[]
@@ -1119,3 +1115,7 @@ type PromptParams = BaseParams & {
 }
 
 export type FetchChatCompletionParams = MessagesParams | PromptParams
+
+// More specific than NonNullable
+export type NotUndefined<T> = Exclude<T, undefined>
+export type NotNull<T> = Exclude<T, null>
diff --git a/src/renderer/src/types/provider.ts b/src/renderer/src/types/provider.ts
index 5bd605007e..05988f6a1f 100644
--- a/src/renderer/src/types/provider.ts
+++ b/src/renderer/src/types/provider.ts
@@ -1,6 +1,9 @@
+import type OpenAI from '@cherrystudio/openai'
 import type { Model } from '@types'
 import * as z from 'zod'
 
+import type { OpenAIVerbosity } from './aiCoreTypes'
+
 export const ProviderTypeSchema = z.enum([
   'openai',
   'openai-response',
@@ -41,36 +44,38 @@ export type ProviderApiOptions = {
   isNotSupportAPIVersion?: boolean
 }
 
+// scale is not well supported now. It even lacks of docs
+// We take undefined as same as default, and null as same as explicitly off.
+// It controls whether the response contains the serviceTier field or not, so undefined and null should be separated.
+export type OpenAIServiceTier = Exclude<OpenAI.Responses.ResponseCreateParams['service_tier'], 'scale'>
+
 export const OpenAIServiceTiers = {
   auto: 'auto',
   default: 'default',
   flex: 'flex',
   priority: 'priority'
-} as const
+} as const satisfies Record<NonNullable<OpenAIServiceTier>, OpenAIServiceTier>
 
-export type OpenAIServiceTier = keyof typeof OpenAIServiceTiers
-
-export function isOpenAIServiceTier(tier: string): tier is OpenAIServiceTier {
-  return Object.hasOwn(OpenAIServiceTiers, tier)
+export function isOpenAIServiceTier(tier: string | null | undefined): tier is OpenAIServiceTier {
+  return tier === null || tier === undefined || Object.hasOwn(OpenAIServiceTiers, tier)
 }
 
+// https://console.groq.com/docs/api-reference#responses
+export type GroqServiceTier = 'auto' | 'on_demand' | 'flex' | undefined | null
+
 export const GroqServiceTiers = {
   auto: 'auto',
   on_demand: 'on_demand',
-  flex: 'flex',
-  performance: 'performance'
-} as const
+  flex: 'flex'
+} as const satisfies Record<string, GroqServiceTier>
 
-// 从 GroqServiceTiers 对象中提取类型
-export type GroqServiceTier = keyof typeof GroqServiceTiers
-
-export function isGroqServiceTier(tier: string): tier is GroqServiceTier {
-  return Object.hasOwn(GroqServiceTiers, tier)
+export function isGroqServiceTier(tier: string | undefined | null): tier is GroqServiceTier {
+  return tier === null || tier === undefined || Object.hasOwn(GroqServiceTiers, tier)
 }
 
 export type ServiceTier = OpenAIServiceTier | GroqServiceTier
 
-export function isServiceTier(tier: string): tier is ServiceTier {
+export function isServiceTier(tier: string | null | undefined): tier is ServiceTier {
   return isGroqServiceTier(tier) || isOpenAIServiceTier(tier)
 }
 
@@ -103,6 +108,7 @@ export type Provider = {
   // API options
   apiOptions?: ProviderApiOptions
   serviceTier?: ServiceTier
+  verbosity?: OpenAIVerbosity
 
   /** @deprecated */
   isNotSupportArrayContent?: boolean
@@ -119,6 +125,75 @@ export type Provider = {
   extra_headers?: Record<string, string>
 }
 
+export const SystemProviderIdSchema = z.enum([
+  'cherryin',
+  'silicon',
+  'aihubmix',
+  'ocoolai',
+  'deepseek',
+  'ppio',
+  'alayanew',
+  'qiniu',
+  'dmxapi',
+  'burncloud',
+  'tokenflux',
+  '302ai',
+  'cephalon',
+  'lanyun',
+  'ph8',
+  'openrouter',
+  'ollama',
+  'ovms',
+  'new-api',
+  'lmstudio',
+  'anthropic',
+  'openai',
+  'azure-openai',
+  'gemini',
+  'vertexai',
+  'github',
+  'copilot',
+  'zhipu',
+  'yi',
+  'moonshot',
+  'baichuan',
+  'dashscope',
+  'stepfun',
+  'doubao',
+  'infini',
+  'minimax',
+  'groq',
+  'together',
+  'fireworks',
+  'nvidia',
+  'grok',
+  'hyperbolic',
+  'mistral',
+  'jina',
+  'perplexity',
+  'modelscope',
+  'xirang',
+  'hunyuan',
+  'tencent-cloud-ti',
+  'baidu-cloud',
+  'gpustack',
+  'voyageai',
+  'aws-bedrock',
+  'poe',
+  'aionly',
+  'longcat',
+  'huggingface',
+  'sophnet',
+  'ai-gateway',
+  'cerebras'
+])
+
+export type SystemProviderId = z.infer<typeof SystemProviderIdSchema>
+
+export const isSystemProviderId = (id: string): id is SystemProviderId => {
+  return SystemProviderIdSchema.safeParse(id).success
+}
+
 export const SystemProviderIds = {
   cherryin: 'cherryin',
   silicon: 'silicon',
@@ -180,13 +255,9 @@ export const SystemProviderIds = {
   huggingface: 'huggingface',
   'ai-gateway': 'ai-gateway',
   cerebras: 'cerebras'
-} as const
+} as const satisfies Record<SystemProviderId, SystemProviderId>
 
-export type SystemProviderId = keyof typeof SystemProviderIds
-
-export const isSystemProviderId = (id: string): id is SystemProviderId => {
-  return Object.hasOwn(SystemProviderIds, id)
-}
+type SystemProviderIdTypeMap = typeof SystemProviderIds
 
 export type SystemProvider = Provider & {
   id: SystemProviderId
@@ -216,3 +287,16 @@ export type AzureOpenAIProvider = Provider & {
 export const isSystemProvider = (provider: Provider): provider is SystemProvider => {
   return isSystemProviderId(provider.id) && !!provider.isSystem
 }
+
+export type GroqSystemProvider = Provider & {
+  id: SystemProviderIdTypeMap['groq']
+  isSystem: true
+}
+
+export type NotGroqProvider = Provider & {
+  id: Exclude<string, SystemProviderIdTypeMap['groq']>
+}
+
+export const isGroqSystemProvider = (provider: Provider): provider is GroqSystemProvider => {
+  return provider.id === SystemProviderIds.groq
+}
diff --git a/src/renderer/src/windows/mini/home/HomeWindow.tsx b/src/renderer/src/windows/mini/home/HomeWindow.tsx
index bf01146015..a3da9d9a0b 100644
--- a/src/renderer/src/windows/mini/home/HomeWindow.tsx
+++ b/src/renderer/src/windows/mini/home/HomeWindow.tsx
@@ -283,7 +283,7 @@ const HomeWindow: FC<{ draggable?: boolean }> = ({ draggable = true }) => {
         await fetchChatCompletion({
           messages: modelMessages,
           assistant: newAssistant,
-          options: {},
+          requestOptions: {},
           topicId,
           uiMessages: uiMessages,
           onChunkReceived: (chunk: Chunk) => {
diff --git a/src/renderer/src/windows/selection/action/components/ActionUtils.ts b/src/renderer/src/windows/selection/action/components/ActionUtils.ts
index 16537f0e81..12f3881fe2 100644
--- a/src/renderer/src/windows/selection/action/components/ActionUtils.ts
+++ b/src/renderer/src/windows/selection/action/components/ActionUtils.ts
@@ -70,7 +70,7 @@ export const processMessages = async (
     await fetchChatCompletion({
       messages: modelMessages,
       assistant: newAssistant,
-      options: {},
+      requestOptions: {},
       uiMessages: uiMessages,
       onChunkReceived: (chunk: Chunk) => {
         if (finished) {