diff --git a/src/common/app.release.ts b/src/common/app.release.ts
index e92272fe2..5d565cc66 100644
--- a/src/common/app.release.ts
+++ b/src/common/app.release.ts
@@ -28,7 +28,7 @@ export const Release = {
 
   // this is here to trigger revalidation of data, e.g. models refresh
   Monotonics: {
-    Aix: 26,
+    Aix: 27,
     NewsVersion: 192,
   },
 
diff --git a/src/common/stores/llms/llms.parameters.ts b/src/common/stores/llms/llms.parameters.ts
index e8813026f..111ca2b5c 100644
--- a/src/common/stores/llms/llms.parameters.ts
+++ b/src/common/stores/llms/llms.parameters.ts
@@ -131,6 +131,14 @@ export const DModelParameterRegistry = {
     initialValue: true,
   } as const,
 
+  llmVndOaiVerbosity: {
+    label: 'Verbosity',
+    type: 'enum' as const,
+    description: 'Controls response length and detail level',
+    values: ['low', 'medium', 'high'] as const,
+    requiredFallback: 'medium',
+  } as const,
+
   llmVndOaiWebSearchContext: {
     label: 'Search Context Size',
     type: 'enum' as const,
diff --git a/src/modules/aix/client/aix.client.ts b/src/modules/aix/client/aix.client.ts
index a83364219..d48490e8b 100644
--- a/src/modules/aix/client/aix.client.ts
+++ b/src/modules/aix/client/aix.client.ts
@@ -49,7 +49,7 @@ export function aixCreateModelFromLLMOptions(
     llmRef, llmTemperature, llmResponseTokens, llmTopP, llmForceNoStream,
     llmVndAntThinkingBudget,
     llmVndGeminiShowThoughts, llmVndGeminiThinkingBudget,
-    llmVndOaiReasoningEffort, llmVndOaiReasoningEffort4, llmVndOaiRestoreMarkdown, llmVndOaiWebSearchContext, llmVndOaiWebSearchGeolocation,
+    llmVndOaiReasoningEffort, llmVndOaiReasoningEffort4, llmVndOaiRestoreMarkdown, llmVndOaiVerbosity, llmVndOaiWebSearchContext, llmVndOaiWebSearchGeolocation,
     llmVndPerplexityDateFilter, llmVndPerplexitySearchMode,
     llmVndXaiSearchMode, llmVndXaiSearchSources, llmVndXaiSearchDateFilter,
   } = {
@@ -105,6 +105,7 @@ export function aixCreateModelFromLLMOptions(
     ...(llmVndOaiResponsesAPI ? { vndOaiResponsesAPI: true } : {}),
     ...((llmVndOaiReasoningEffort4 || llmVndOaiReasoningEffort) ? { vndOaiReasoningEffort: llmVndOaiReasoningEffort4 || llmVndOaiReasoningEffort } : {}),
     ...(llmVndOaiRestoreMarkdown ? { vndOaiRestoreMarkdown: llmVndOaiRestoreMarkdown } : {}),
+    ...(llmVndOaiVerbosity ? { vndOaiVerbosity: llmVndOaiVerbosity } : {}),
     ...(llmVndOaiWebSearchContext ? { vndOaiWebSearchContext: llmVndOaiWebSearchContext } : {}),
     ...(llmVndPerplexityDateFilter ? { vndPerplexityDateFilter: llmVndPerplexityDateFilter } : {}),
     ...(llmVndPerplexitySearchMode ? { vndPerplexitySearchMode: llmVndPerplexitySearchMode } : {}),
diff --git a/src/modules/aix/server/api/aix.wiretypes.ts b/src/modules/aix/server/api/aix.wiretypes.ts
index de7c39327..d9c3ff4ae 100644
--- a/src/modules/aix/server/api/aix.wiretypes.ts
+++ b/src/modules/aix/server/api/aix.wiretypes.ts
@@ -410,6 +410,7 @@ export namespace AixWire_API {
     vndOaiResponsesAPI: z.boolean().optional(),
     vndOaiReasoningEffort: z.enum(['minimal', 'low', 'medium', 'high']).optional(),
     vndOaiRestoreMarkdown: z.boolean().optional(),
+    vndOaiVerbosity: z.enum(['low', 'medium', 'high']).optional(),
     vndOaiWebSearchContext: z.enum(['low', 'medium', 'high']).optional(),
     vndPerplexityDateFilter: z.enum(['unfiltered', '1m', '3m', '6m', '1y']).optional(),
     vndPerplexitySearchMode: z.enum(['default', 'academic']).optional(),
diff --git a/src/modules/aix/server/dispatch/chatGenerate/adapters/openai.responsesCreate.ts b/src/modules/aix/server/dispatch/chatGenerate/adapters/openai.responsesCreate.ts
index 5d639d5ce..8dee853ee 100644
--- a/src/modules/aix/server/dispatch/chatGenerate/adapters/openai.responsesCreate.ts
+++ b/src/modules/aix/server/dispatch/chatGenerate/adapters/openai.responsesCreate.ts
@@ -93,6 +93,14 @@ export function aixToOpenAIResponses(model: AixAPI_Model, chatGenerate: AixAPICh
     // };
   }
 
+  // GPT-5 Verbosity: Add to existing text config or create new one
+  if (model.vndOaiVerbosity) {
+    payload.text = {
+      ...payload.text,
+      verbosity: model.vndOaiVerbosity,
+    };
+  }
+
   // Tool: Search: for search models, and deep research models
   // NOTE: OpenAI doesn't support web search with minimal reasoning effort
   const skipWebSearchDueToMinimalReasoning = model.vndOaiReasoningEffort === 'minimal';
diff --git a/src/modules/aix/server/dispatch/wiretypes/openai.wiretypes.ts b/src/modules/aix/server/dispatch/wiretypes/openai.wiretypes.ts
index 643235db4..ca894c080 100644
--- a/src/modules/aix/server/dispatch/wiretypes/openai.wiretypes.ts
+++ b/src/modules/aix/server/dispatch/wiretypes/openai.wiretypes.ts
@@ -1202,6 +1202,7 @@ export namespace OpenAIWire_API_Responses {
         }),
         // z.object({ type: z.literal('json_object') }), // deprecated
       ]).optional(),
+      verbosity: z.enum(['low', 'medium', 'high']).optional(), // GPT-5 verbosity control
     }).optional(),
 
     // State management (we won't use this for stateless)
diff --git a/src/modules/llms/server/llm.server.types.ts b/src/modules/llms/server/llm.server.types.ts
index 7eeedba7a..1e620c1b9 100644
--- a/src/modules/llms/server/llm.server.types.ts
+++ b/src/modules/llms/server/llm.server.types.ts
@@ -81,6 +81,7 @@ const ModelParameterSpec_schema = z.object({
     'llmVndOaiReasoningEffort',
     'llmVndOaiReasoningEffort4',
     'llmVndOaiRestoreMarkdown',
+    'llmVndOaiVerbosity',
     'llmVndOaiWebSearchContext',
     'llmVndOaiWebSearchGeolocation',
     'llmVndPerplexityDateFilter',