Support configure thinking parameters for Anthropic and adjust background color for outline text field

desktop-shared/src/main/kotlin/io/askimo/ui/common/theme/AppComponents.kt

@@ -230,6 +230,7 @@ object AppComponents {
         focusedTextColor: Color = MaterialTheme.colorScheme.onSurface,
         unfocusedTextColor: Color = MaterialTheme.colorScheme.onSurface,
         cursorColor: Color = MaterialTheme.colorScheme.onSurface,
+        containerColor: Color = MaterialTheme.colorScheme.surface,
     ): TextFieldColors = OutlinedTextFieldDefaults.colors(
         focusedBorderColor = focusedBorderColor,
         unfocusedBorderColor = unfocusedBorderColor,
@@ -238,6 +239,9 @@ object AppComponents {
         focusedTextColor = focusedTextColor,
         unfocusedTextColor = unfocusedTextColor,
         cursorColor = cursorColor,
+        focusedContainerColor = containerColor,
+        unfocusedContainerColor = containerColor,
+        disabledContainerColor = containerColor,
     )
 
     /**

desktop-shared/src/main/resources/i18n/messages.properties

@@ -949,6 +949,12 @@ provider.openai.setup.help=💡To use OpenAI, you need to provide an API key.\n\
 provider.anthropic.apikey.description=Your Anthropic API key from https://console.anthropic.com/
 provider.anthropic.apikey.stored=API key already stored securely. Leave blank to keep existing key, or enter a new one to update.
 provider.anthropic.setup.help=💡To use Anthropic, you need to provide an API key.\n\n1. Get your API key from: https://console.anthropic.com/\n2. Then set it in the Settings or using: :change-settings
+provider.anthropic.max_tokens.label=Max Tokens
+provider.anthropic.max_tokens.description=Maximum tokens the model can generate per response (thinking + answer combined).
+provider.anthropic.thinking_budget_tokens.label=Thinking Budget Tokens
+provider.anthropic.thinking_budget_tokens.description=Max tokens Anthropic can spend on thinking. Set to 0 to let the model decide automatically (recommended).
+provider.anthropic.thinking_max_tokens.label=Thinking Max Tokens
+provider.anthropic.thinking_max_tokens.description=Total token cap when thinking is active (thinking + answer). Set to 0 to use Max Tokens instead.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Your Google AI Studio API key from https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_de.properties

@@ -909,6 +909,12 @@ provider.openai.setup.help=💡 Für die Nutzung von OpenAI benötigen Sie einen
 provider.anthropic.apikey.description=Ihr Anthropic API-Schlüssel: https://console.anthropic.com/
 provider.anthropic.apikey.stored=API-Schlüssel ist bereits gespeichert.
 provider.anthropic.setup.help=💡 Für die Nutzung von Anthropic benötigen Sie einen API-Schlüssel.\n\n1. Hier erhalten: https://console.anthropic.com/\n2. In den Einstellungen eintragen oder :change-settings verwenden
+provider.anthropic.max_tokens.label=Max Tokens
+provider.anthropic.max_tokens.description=Maximale Anzahl an Tokens, die das Modell pro Antwort generieren kann (Denken + Antwort kombiniert).
+provider.anthropic.thinking_budget_tokens.label=Thinking Budget Tokens
+provider.anthropic.thinking_budget_tokens.description=Maximale Anzahl an Tokens, die Anthropic für den Denkprozess verwenden kann. Auf 0 setzen, damit das Modell dies automatisch entscheidet (empfohlen).
+provider.anthropic.thinking_max_tokens.label=Thinking Max Tokens
+provider.anthropic.thinking_max_tokens.description=Gesamtes Token-Limit bei aktivem Denkprozess (Denken + Antwort). Auf 0 setzen, um stattdessen "Max Tokens" zu verwenden.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Ihr Google AI Studio API-Schlüssel: https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_es.properties

@@ -907,6 +907,12 @@ provider.openai.setup.help=💡 Para usar OpenAI necesitas una clave API.\n\n1.
 provider.anthropic.apikey.description=Tu clave API de Anthropic: https://console.anthropic.com/
 provider.anthropic.apikey.stored=Clave API ya almacenada.
 provider.anthropic.setup.help=💡 Para usar Anthropic, necesitas una clave API.\n\n1. Consíguela en: https://console.anthropic.com/\n2. Configúrala en los ajustes o con :change-settings
+provider.anthropic.max_tokens.label=Max Tokens
+provider.anthropic.max_tokens.description=Máximo de tokens que el modelo puede generar por respuesta (pensamiento + respuesta combinados).
+provider.anthropic.thinking_budget_tokens.label=Tokens de presupuesto de pensamiento
+provider.anthropic.thinking_budget_tokens.description=Máximo de tokens que Anthropic puede dedicar al pensamiento. Configura en 0 para dejar que el modelo decida automáticamente (recomendado).
+provider.anthropic.thinking_max_tokens.label=Máximo de tokens de pensamiento
+provider.anthropic.thinking_max_tokens.description=Límite total de tokens cuando el pensamiento está activo (pensamiento + respuesta). Configura en 0 para usar Max Tokens en su lugar.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Tu clave API de Google AI Studio: https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_fr.properties

@@ -907,6 +907,12 @@ provider.openai.setup.help=💡 Pour utiliser OpenAI, une clé API est nécessai
 provider.anthropic.apikey.description=Votre clé API Anthropic : https://console.anthropic.com/
 provider.anthropic.apikey.stored=Clé API déjà stockée.
 provider.anthropic.setup.help=💡 Pour utiliser Anthropic, une clé API est nécessaire.\n\n1. Obtenez-la ici : https://console.anthropic.com/\n2. Configurez-la dans les paramètres ou via :change-settings
+provider.anthropic.max_tokens.label=Max Tokens
+provider.anthropic.max_tokens.description=Nombre maximal de jetons que le modèle peut générer par réponse (réflexion + réponse combinées).
+provider.anthropic.thinking_budget_tokens.label=Jetons de budget de réflexion
+provider.anthropic.thinking_budget_tokens.description=Nombre maximal de jetons qu'Anthropic peut consacrer à la réflexion. Définissez sur 0 pour laisser le modèle décider automatiquement (recommandé).
+provider.anthropic.thinking_max_tokens.label=Max de jetons de réflexion
+provider.anthropic.thinking_max_tokens.description=Plafond total de jetons lorsque la réflexion est activée (réflexion + réponse). Définissez sur 0 pour utiliser Max Tokens à la place.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Votre clé API Google AI Studio : https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_ja_JP.properties

@@ -907,6 +907,12 @@ provider.openai.setup.help=💡 OpenAI を使うには API キーが必要です
 provider.anthropic.apikey.description=Anthropic API キー: https://console.anthropic.com/
 provider.anthropic.apikey.stored=API キーは既に保存されています。
 provider.anthropic.setup.help=💡 Anthropic を使うには API キーが必要です。\n\n1. https://console.anthropic.com/ で取得\n2. 設定で構成するか :change-settings を使用
+provider.anthropic.max_tokens.label=最大トークン数
+provider.anthropic.max_tokens.description=モデルが応答ごとに生成できる最大トークン数（思考 + 回答の合計）。
+provider.anthropic.thinking_budget_tokens.label=思考予算トークン数
+provider.anthropic.thinking_budget_tokens.description=Anthropicが思考に費やすことのできる最大トークン数。0に設定すると、モデルが自動的に決定します（推奨）。
+provider.anthropic.thinking_max_tokens.label=思考時最大トークン数
+provider.anthropic.thinking_max_tokens.description=思考がアクティブな時の合計トークン上限（思考 + 回答）。0に設定すると、代わりに「Max Tokens」が使用されます。
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Google AI Studio API キー: https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_ko_KR.properties

@@ -907,6 +907,12 @@ provider.openai.setup.help=💡 OpenAI 사용을 위해 API 키가 필요합니
 provider.anthropic.apikey.description=Anthropic API 키: https://console.anthropic.com/
 provider.anthropic.apikey.stored=API 키가 이미 저장됨.
 provider.anthropic.setup.help=💡 Anthropic 사용을 위해 API 키가 필요합니다.\n\n1. https://console.anthropic.com/ 에서 키를 받으세요\n2. 설정 또는 :change-settings 명령으로 구성하세요
+provider.anthropic.max_tokens.label=최대 토큰 수
+provider.anthropic.max_tokens.description=모델이 응답당 생성할 수 있는 최대 토큰 수 (생각 + 답변 합계).
+provider.anthropic.thinking_budget_tokens.label=생각 예산 토큰 수
+provider.anthropic.thinking_budget_tokens.description=Anthropic이 생각하는 데 사용할 수 있는 최대 토큰 수입니다. 0으로 설정하면 모델이 자동으로 결정합니다 (권장).
+provider.anthropic.thinking_max_tokens.label=생각 최대 토큰 수
+provider.anthropic.thinking_max_tokens.description=생각 기능이 활성화되었을 때의 총 토큰 제한 (생각 + 답변). 0으로 설정하면 대신 최대 토큰 수(Max Tokens)를 사용합니다.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Google AI Studio API 키: https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_pt_BR.properties

@@ -910,6 +910,12 @@ provider.openai.setup.help=💡 Para usar o OpenAI, forneça sua API Key.\n\n1.
 provider.anthropic.apikey.description=Sua API Key do Anthropic: https://console.anthropic.com/
 provider.anthropic.apikey.stored=API Key já armazenada.
 provider.anthropic.setup.help=💡 Para usar o Anthropic, forneça a API Key.\n\n1. Obtenha em: https://console.anthropic.com/\n2. Configure nas Configurações ou usando :change-settings
+provider.anthropic.max_tokens.label=Max Tokens
+provider.anthropic.max_tokens.description=Máximo de tokens que o modelo pode gerar por resposta (pensamento + resposta combinados).
+provider.anthropic.thinking_budget_tokens.label=Tokens de orçamento de pensamento
+provider.anthropic.thinking_budget_tokens.description=Máximo de tokens que a Anthropic pode gastar pensando. Defina como 0 para permitir que o modelo decida automaticamente (recomendado).
+provider.anthropic.thinking_max_tokens.label=Máximo de tokens de pensamento
+provider.anthropic.thinking_max_tokens.description=Limite total de tokens quando o pensamento está ativo (pensamento + resposta). Defina como 0 para usar Max Tokens em vez disso.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Sua API Key do Google AI Studio: https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_vi_VN.properties

@@ -906,6 +906,12 @@ provider.openai.setup.help=💡 Để sử dụng OpenAI, bạn cần API key.\n
 provider.anthropic.apikey.description=API key Anthropic từ https://console.anthropic.com/
 provider.anthropic.apikey.stored=API key đã lưu. Để trống để giữ nguyên hoặc nhập mới để cập nhật.
 provider.anthropic.setup.help=💡 Để sử dụng Anthropic, bạn cần API key.\n\n1. Lấy từ https://console.anthropic.com/\n2. Thiết lập trong Cài đặt hoặc dùng :change-settings
+provider.anthropic.max_tokens.label=Số Token Tối đa
+provider.anthropic.max_tokens.description=Số lượng token tối đa mà mô hình có thể tạo cho mỗi phản hồi (suy nghĩ + câu trả lời được kết hợp).
+provider.anthropic.thinking_budget_tokens.label=Token Ngân sách Suy nghĩ
+provider.anthropic.thinking_budget_tokens.description=Số lượng token tối đa mà Anthropic có thể dành cho việc suy nghĩ. Đặt là 0 để để mô hình tự quyết định (khuyên dùng).
+provider.anthropic.thinking_max_tokens.label=Token Tối đa cho Suy nghĩ
+provider.anthropic.thinking_max_tokens.description=Giới hạn token tổng khi tính năng suy nghĩ đang hoạt động (suy nghĩ + câu trả lời). Đặt là 0 để sử dụng Số Token Tối đa thay thế.
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=API key Google AI Studio từ https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_zh_CN.properties

@@ -907,6 +907,12 @@ provider.openai.setup.help=使用 OpenAI 需要 API Key。\n\n1. 获取：https:
 provider.anthropic.apikey.description=Anthropic API Key：https://console.anthropic.com/
 provider.anthropic.apikey.stored=API Key 已保存。
 provider.anthropic.setup.help=使用 Anthropic 需要 API Key。\n\n1. 获取：https://console.anthropic.com/\n2. 在设置中配置或使用 :change-settings
+provider.anthropic.max_tokens.label=最大 Token 数
+provider.anthropic.max_tokens.description=模型每次响应可生成的最大 Token 数（思考 + 回答的总和）。
+provider.anthropic.thinking_budget_tokens.label=思考预算 Token 数
+provider.anthropic.thinking_budget_tokens.description=Anthropic 用于思考的最大 Token 数。设置为 0 可让模型自动决定（推荐）。
+provider.anthropic.thinking_max_tokens.label=思考时最大 Token 数
+provider.anthropic.thinking_max_tokens.description=启用思考功能时的总 Token 上限（思考 + 回答）。设置为 0 以使用“最大 Token 数”代替。
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=Google AI Studio API Key：https://aistudio.google.com/app/apikey

desktop-shared/src/main/resources/i18n/messages_zh_TW.properties

@@ -907,6 +907,12 @@ provider.openai.setup.help=💡 若要使用 OpenAI，您必須提供 API 金鑰
 provider.anthropic.apikey.description=您的 Anthropic API 金鑰（https://console.anthropic.com/）
 provider.anthropic.apikey.stored=API 金鑰已儲存。留白代表保持現有設定。
 provider.anthropic.setup.help=💡 若要使用 Anthropic，請提供 API 金鑰。\n\n1. 取得位置：https://console.anthropic.com/\n2. 在設定中輸入或使用 :change-settings
+provider.anthropic.max_tokens.label=最大 Token 數
+provider.anthropic.max_tokens.description=模型每次回應可生成的最大 Token 數（思考 + 回答的總和）。
+provider.anthropic.thinking_budget_tokens.label=思考預算 Token 數
+provider.anthropic.thinking_budget_tokens.description=Anthropic 用於思考的最大 Token 數。設定為 0 可讓模型自動決定（推薦）。
+provider.anthropic.thinking_max_tokens.label=思考時最大 Token 數
+provider.anthropic.thinking_max_tokens.description=啟用思考功能時的總 Token 上限（思考 + 回答）。設定為 0 以使用「最大 Token 數」代替。
 
 # Provider Configuration - Gemini
 provider.gemini.apikey.description=您的 Google AI Studio API 金鑰（https://aistudio.google.com/app/apikey）

desktop/src/main/kotlin/io/askimo/desktop/settings/AIProviderSettingsSection.kt

@@ -19,6 +19,7 @@ import androidx.compose.foundation.layout.width
 import androidx.compose.foundation.layout.widthIn
 import androidx.compose.foundation.rememberScrollState
 import androidx.compose.foundation.rememberScrollbarAdapter
+import androidx.compose.foundation.text.KeyboardOptions
 import androidx.compose.foundation.verticalScroll
 import androidx.compose.material.icons.Icons
 import androidx.compose.material.icons.filled.Add
@@ -41,9 +42,11 @@ import androidx.compose.runtime.remember
 import androidx.compose.runtime.setValue
 import androidx.compose.ui.Alignment
 import androidx.compose.ui.Modifier
+import androidx.compose.ui.text.input.KeyboardType
 import androidx.compose.ui.text.style.TextOverflow
 import androidx.compose.ui.unit.dp
 import io.askimo.core.AppConstants.DOMAIN
+import io.askimo.core.i18n.LocalizationManager
 import io.askimo.core.providers.ChatModelFactory
 import io.askimo.core.providers.ModelDTO
 import io.askimo.core.providers.ModelProvider
@@ -317,6 +320,20 @@ private fun providerModelConfigCard(instance: ProviderInstance, viewModel: AIPro
                 )
             }
 
+            // Provider-specific configurable fields (e.g. maxTokens for Anthropic)
+            val configurableFields = instance.settings.getConfigurableFields { key -> LocalizationManager.getString(key) }
+            if (configurableFields.isNotEmpty()) {
+                HorizontalDivider(color = MaterialTheme.colorScheme.onSecondaryContainer.copy(alpha = 0.15f))
+                configurableFields.forEach { field ->
+                    providerConfigurableField(
+                        field = field,
+                        onValueChange = { newValue ->
+                            viewModel.updateInstanceModelOverride(instance.id, field.name, newValue)
+                        },
+                    )
+                }
+            }
+
             // Embedding model — only for supported providers
             if (supportsEmbedding) {
                 HorizontalDivider(color = MaterialTheme.colorScheme.onSecondaryContainer.copy(alpha = 0.15f))
@@ -591,3 +608,73 @@ private fun providerModelTypePickerDialog(
         }
     }
 }
+
+/**
+ * Renders a single provider-specific configurable field (NumberField or TextField).
+ * Called generically for every entry in [ProviderSettings.getConfigurableFields] —
+ * no per-provider branching needed in the UI.
+ */
+@Composable
+private fun providerConfigurableField(
+    field: SettingField,
+    onValueChange: (String) -> Unit,
+) {
+    Row(
+        modifier = Modifier.fillMaxWidth(),
+        horizontalArrangement = Arrangement.SpaceBetween,
+        verticalAlignment = Alignment.CenterVertically,
+    ) {
+        Column(
+            modifier = Modifier.weight(1f).padding(end = Spacing.large),
+            verticalArrangement = Arrangement.spacedBy(Spacing.extraSmall),
+        ) {
+            Text(
+                text = field.label,
+                style = MaterialTheme.typography.bodyMedium,
+                color = MaterialTheme.colorScheme.onSecondaryContainer,
+            )
+            Text(
+                text = field.description,
+                style = MaterialTheme.typography.bodySmall,
+                color = MaterialTheme.colorScheme.onSecondaryContainer.copy(alpha = 0.6f),
+            )
+        }
+
+        when (field) {
+            is SettingField.NumberField -> {
+                var text by remember(field.value) { mutableStateOf(field.value.toString()) }
+                OutlinedTextField(
+                    value = text,
+                    onValueChange = { newVal ->
+                        text = newVal
+                        newVal.toIntOrNull()?.let { onValueChange(it.toString()) }
+                    },
+                    modifier = Modifier.widthIn(min = 100.dp, max = 160.dp),
+                    singleLine = true,
+                    keyboardOptions = KeyboardOptions(keyboardType = KeyboardType.Number),
+                    colors = AppComponents.outlinedTextFieldColors(
+                        containerColor = MaterialTheme.colorScheme.surface,
+                    ),
+                )
+            }
+
+            is SettingField.TextField -> {
+                var text by remember(field.value) { mutableStateOf(field.value) }
+                OutlinedTextField(
+                    value = text,
+                    onValueChange = { newVal ->
+                        text = newVal
+                        onValueChange(newVal)
+                    },
+                    modifier = Modifier.widthIn(min = 100.dp, max = 200.dp),
+                    singleLine = true,
+                    colors = AppComponents.outlinedTextFieldColors(
+                        containerColor = MaterialTheme.colorScheme.surface,
+                    ),
+                )
+            }
+
+            else -> Unit
+        }
+    }
+}

desktop/src/main/kotlin/io/askimo/desktop/settings/AdvancedSettingsSection.kt

@@ -179,9 +179,7 @@ private fun logLevelCard() {
                             modifier = Modifier
                                 .fillMaxWidth()
                                 .clickableCard { logLevelDropdownExpanded = true },
-                            colors = CardDefaults.cardColors(
-                                containerColor = MaterialTheme.colorScheme.surface,
-                            ),
+                            colors = CardDefaults.cardColors(),
                         ) {
                             Row(
                                 modifier = Modifier