diff --git a/.github/scripts/autogen/types.ts b/.github/scripts/autogen/types.ts
index 261723b67..a5ac6767c 100644
--- a/.github/scripts/autogen/types.ts
+++ b/.github/scripts/autogen/types.ts
@@ -157,6 +157,12 @@ export interface components {
             supportedModes?: components["schemas"]["Mode"][];
             /** @description Whether the model supports extended thinking / reasoning */
             thinking?: boolean;
+            /**
+             * @description When true, inbound Chat Completions stay on /chat/completions unless a
+             *     responses-only model or an existing trigger (header / reasoning / Cursor).
+             *     Omit on new models so Chat Completions is always served via Responses.
+             */
+            preferChatCompletions?: boolean;
         };
         ModelParam: {
             defaultValue?: (string | number | boolean) | null;
diff --git a/.github/test/model.cue b/.github/test/model.cue
index 497466c9a..e334b0fa9 100644
--- a/.github/test/model.cue
+++ b/.github/test/model.cue
@@ -341,6 +341,10 @@ package model
 	supportedModes?: [...#Mode]
 	// Whether the model supports extended thinking / reasoning
 	thinking?: bool
+	// When true, inbound Chat Completions stay on /chat/completions unless a
+	// responses-only model or an existing trigger (header / reasoning / Cursor).
+	// Omit on new models so Chat Completions is always served via Responses.
+	preferChatCompletions?: bool
 }
 
 #PricingTier: {
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index d47213744..34ca54175 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -147,6 +147,14 @@ removeParams: [<param-key>, ...]
 # Param keys that must always be provided by callers
 requiredParams: [<param-key>, ...]
 
+# Whether the model supports extended thinking / reasoning
+thinking: true
+
+# When true, inbound Chat Completions stay on /chat/completions unless a
+# responses-only model or an existing trigger (header / reasoning / Cursor).
+# Omit on new models so Chat Completions is always served via Responses.
+preferChatCompletions: true
+
 # Provider-specific escape-hatch configuration. Only set a key here when the
 # behaviour cannot be expressed through the generic fields above, and only on
 # models belonging to the provider that owns the key.
diff --git a/providers/aws-bedrock-mantle/google.gemma-4-26b-a4b.yaml b/providers/aws-bedrock-mantle/google.gemma-4-26b-a4b.yaml
index 15255e5b2..7467bdc78 100644
--- a/providers/aws-bedrock-mantle/google.gemma-4-26b-a4b.yaml
+++ b/providers/aws-bedrock-mantle/google.gemma-4-26b-a4b.yaml
@@ -30,6 +30,7 @@ modalities:
         - text
 mode: chat
 model: google.gemma-4-26b-a4b
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-google-gemma-4-26b-a4b.html
diff --git a/providers/aws-bedrock-mantle/google.gemma-4-31b.yaml b/providers/aws-bedrock-mantle/google.gemma-4-31b.yaml
index a4967808d..67de176f3 100644
--- a/providers/aws-bedrock-mantle/google.gemma-4-31b.yaml
+++ b/providers/aws-bedrock-mantle/google.gemma-4-31b.yaml
@@ -27,6 +27,7 @@ modalities:
         - text
 mode: chat
 model: google.gemma-4-31b
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-google-gemma-4-31b.html
diff --git a/providers/aws-bedrock-mantle/google.gemma-4-e2b.yaml b/providers/aws-bedrock-mantle/google.gemma-4-e2b.yaml
index 68d305602..a1ec7e6c8 100644
--- a/providers/aws-bedrock-mantle/google.gemma-4-e2b.yaml
+++ b/providers/aws-bedrock-mantle/google.gemma-4-e2b.yaml
@@ -28,6 +28,7 @@ modalities:
         - text
 mode: chat
 model: google.gemma-4-e2b
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-google-gemma-4-e2b.html
diff --git a/providers/aws-bedrock-mantle/openai.gpt-oss-120b.yaml b/providers/aws-bedrock-mantle/openai.gpt-oss-120b.yaml
index 76eafc10f..73d29fe99 100644
--- a/providers/aws-bedrock-mantle/openai.gpt-oss-120b.yaml
+++ b/providers/aws-bedrock-mantle/openai.gpt-oss-120b.yaml
@@ -69,6 +69,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-oss-120b.html
diff --git a/providers/aws-bedrock-mantle/openai.gpt-oss-20b.yaml b/providers/aws-bedrock-mantle/openai.gpt-oss-20b.yaml
index c899cd269..6cc35ecef 100644
--- a/providers/aws-bedrock-mantle/openai.gpt-oss-20b.yaml
+++ b/providers/aws-bedrock-mantle/openai.gpt-oss-20b.yaml
@@ -61,6 +61,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-oss-20b.html
diff --git a/providers/aws-bedrock-mantle/xai.grok-4.3.yaml b/providers/aws-bedrock-mantle/xai.grok-4.3.yaml
index 4410f69ac..053cbf0c1 100644
--- a/providers/aws-bedrock-mantle/xai.grok-4.3.yaml
+++ b/providers/aws-bedrock-mantle/xai.grok-4.3.yaml
@@ -51,6 +51,7 @@ params:
     - defaultValue: 131072
       key: max_completion_tokens
       type: number
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-xai-grok-4-3.html
diff --git a/providers/aws-bedrock-mantle/xai.grok-4.6.yaml b/providers/aws-bedrock-mantle/xai.grok-4.6.yaml
index b2cb656d7..baa89b7c0 100644
--- a/providers/aws-bedrock-mantle/xai.grok-4.6.yaml
+++ b/providers/aws-bedrock-mantle/xai.grok-4.6.yaml
@@ -39,6 +39,7 @@ params:
     - defaultValue: 131072
       key: max_completion_tokens
       type: number
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-xai-grok-4-6.html
diff --git a/providers/aws-bedrock/openai.gpt-oss-120b-1:0.yaml b/providers/aws-bedrock/openai.gpt-oss-120b-1:0.yaml
index 45c5be4f7..dd352e19c 100644
--- a/providers/aws-bedrock/openai.gpt-oss-120b-1:0.yaml
+++ b/providers/aws-bedrock/openai.gpt-oss-120b-1:0.yaml
@@ -84,5 +84,4 @@ sources:
 status: active
 supportedModes:
     - chat
-    - responses
 thinking: true
diff --git a/providers/aws-bedrock/openai.gpt-oss-20b-1:0.yaml b/providers/aws-bedrock/openai.gpt-oss-20b-1:0.yaml
index 5e38f705e..f0ca87943 100644
--- a/providers/aws-bedrock/openai.gpt-oss-20b-1:0.yaml
+++ b/providers/aws-bedrock/openai.gpt-oss-20b-1:0.yaml
@@ -85,5 +85,4 @@ sources:
 status: active
 supportedModes:
     - chat
-    - responses
 thinking: true
diff --git a/providers/azure-open-ai/computer-use-preview-2025-04-15.yaml b/providers/azure-open-ai/computer-use-preview-2025-04-15.yaml
index a52354f3a..6f0a01a2f 100644
--- a/providers/azure-open-ai/computer-use-preview-2025-04-15.yaml
+++ b/providers/azure-open-ai/computer-use-preview-2025-04-15.yaml
@@ -27,6 +27,7 @@ params:
       key: max_tokens
       maxValue: 1024
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 status: deprecated
 supportedModes:
diff --git a/providers/azure-open-ai/gpt-4.1-2025-04-14.yaml b/providers/azure-open-ai/gpt-4.1-2025-04-14.yaml
index 829aee5b9..460b34eaa 100644
--- a/providers/azure-open-ai/gpt-4.1-2025-04-14.yaml
+++ b/providers/azure-open-ai/gpt-4.1-2025-04-14.yaml
@@ -35,6 +35,7 @@ model: gpt-4.1-2025-04-14
 params:
     - key: max_tokens
       maxValue: 32768
+preferChatCompletions: true
 retirementDate: "2027-04-14"
 sources:
     - https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/models
diff --git a/providers/azure-open-ai/gpt-4.1-mini-2025-04-14.yaml b/providers/azure-open-ai/gpt-4.1-mini-2025-04-14.yaml
index bb1d0ba01..cef9c1e02 100644
--- a/providers/azure-open-ai/gpt-4.1-mini-2025-04-14.yaml
+++ b/providers/azure-open-ai/gpt-4.1-mini-2025-04-14.yaml
@@ -35,6 +35,7 @@ model: gpt-4.1-mini-2025-04-14
 params:
     - key: max_tokens
       maxValue: 32768
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 sources:
diff --git a/providers/azure-open-ai/gpt-4.1-mini.yaml b/providers/azure-open-ai/gpt-4.1-mini.yaml
index 63b769cfe..9fb9a0e91 100644
--- a/providers/azure-open-ai/gpt-4.1-mini.yaml
+++ b/providers/azure-open-ai/gpt-4.1-mini.yaml
@@ -37,6 +37,7 @@ params:
       key: max_completion_tokens
       maxValue: 32768
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-4.1-nano-2025-04-14.yaml b/providers/azure-open-ai/gpt-4.1-nano-2025-04-14.yaml
index 667f610f2..4ae7989fa 100644
--- a/providers/azure-open-ai/gpt-4.1-nano-2025-04-14.yaml
+++ b/providers/azure-open-ai/gpt-4.1-nano-2025-04-14.yaml
@@ -37,6 +37,7 @@ params:
       key: max_tokens
       maxValue: 32768
       minValue: 1
+preferChatCompletions: true
 removeParams:
     - reasoning_effort
 retirementDate: "2027-04-14"
diff --git a/providers/azure-open-ai/gpt-4.1-nano.yaml b/providers/azure-open-ai/gpt-4.1-nano.yaml
index d78c977cb..07a9de74e 100644
--- a/providers/azure-open-ai/gpt-4.1-nano.yaml
+++ b/providers/azure-open-ai/gpt-4.1-nano.yaml
@@ -40,6 +40,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-4.1.yaml b/providers/azure-open-ai/gpt-4.1.yaml
index 9ca3826a8..5cca971df 100644
--- a/providers/azure-open-ai/gpt-4.1.yaml
+++ b/providers/azure-open-ai/gpt-4.1.yaml
@@ -46,6 +46,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-4o-2024-05-13.yaml b/providers/azure-open-ai/gpt-4o-2024-05-13.yaml
index da69ac3aa..0939eed7a 100644
--- a/providers/azure-open-ai/gpt-4o-2024-05-13.yaml
+++ b/providers/azure-open-ai/gpt-4o-2024-05-13.yaml
@@ -23,6 +23,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/azure-open-ai/gpt-4o-2024-08-06.yaml b/providers/azure-open-ai/gpt-4o-2024-08-06.yaml
index 5ba2bc10c..70ed82f11 100644
--- a/providers/azure-open-ai/gpt-4o-2024-08-06.yaml
+++ b/providers/azure-open-ai/gpt-4o-2024-08-06.yaml
@@ -33,6 +33,7 @@ params:
       key: max_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 sources:
     - https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/models
 status: deprecated
diff --git a/providers/azure-open-ai/gpt-4o-2024-11-20.yaml b/providers/azure-open-ai/gpt-4o-2024-11-20.yaml
index c62cb8272..4181063ea 100644
--- a/providers/azure-open-ai/gpt-4o-2024-11-20.yaml
+++ b/providers/azure-open-ai/gpt-4o-2024-11-20.yaml
@@ -39,6 +39,7 @@ model: gpt-4o-2024-11-20
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/azure-open-ai/gpt-4o-mini-2024-07-18.yaml b/providers/azure-open-ai/gpt-4o-mini-2024-07-18.yaml
index 36748a81e..43bc2ca54 100644
--- a/providers/azure-open-ai/gpt-4o-mini-2024-07-18.yaml
+++ b/providers/azure-open-ai/gpt-4o-mini-2024-07-18.yaml
@@ -38,6 +38,7 @@ model: gpt-4o-mini-2024-07-18
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 status: deprecated
diff --git a/providers/azure-open-ai/gpt-4o-mini.yaml b/providers/azure-open-ai/gpt-4o-mini.yaml
index 1464a6faa..b12e5ba13 100644
--- a/providers/azure-open-ai/gpt-4o-mini.yaml
+++ b/providers/azure-open-ai/gpt-4o-mini.yaml
@@ -40,6 +40,7 @@ params:
       key: max_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 sources:
diff --git a/providers/azure-open-ai/gpt-4o.yaml b/providers/azure-open-ai/gpt-4o.yaml
index d311471f9..26e0e5262 100644
--- a/providers/azure-open-ai/gpt-4o.yaml
+++ b/providers/azure-open-ai/gpt-4o.yaml
@@ -35,6 +35,7 @@ model: gpt-4o
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-10-01"
 status: deprecated
diff --git a/providers/azure-open-ai/gpt-5-2025-08-07.yaml b/providers/azure-open-ai/gpt-5-2025-08-07.yaml
index e37ca5d0a..fcf34df74 100644
--- a/providers/azure-open-ai/gpt-5-2025-08-07.yaml
+++ b/providers/azure-open-ai/gpt-5-2025-08-07.yaml
@@ -49,6 +49,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-chat-2025-08-07.yaml b/providers/azure-open-ai/gpt-5-chat-2025-08-07.yaml
index 3c1182347..ffd47c2b3 100644
--- a/providers/azure-open-ai/gpt-5-chat-2025-08-07.yaml
+++ b/providers/azure-open-ai/gpt-5-chat-2025-08-07.yaml
@@ -31,6 +31,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 removeParams:
     - max_tokens
 sources:
diff --git a/providers/azure-open-ai/gpt-5-chat-2025-10-03.yaml b/providers/azure-open-ai/gpt-5-chat-2025-10-03.yaml
index b69da4b74..127ba14b6 100644
--- a/providers/azure-open-ai/gpt-5-chat-2025-10-03.yaml
+++ b/providers/azure-open-ai/gpt-5-chat-2025-10-03.yaml
@@ -28,6 +28,7 @@ model: gpt-5-chat-2025-10-03
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-chat-latest.yaml b/providers/azure-open-ai/gpt-5-chat-latest.yaml
index f9f390f41..30dcb6bae 100644
--- a/providers/azure-open-ai/gpt-5-chat-latest.yaml
+++ b/providers/azure-open-ai/gpt-5-chat-latest.yaml
@@ -29,6 +29,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-chat.yaml b/providers/azure-open-ai/gpt-5-chat.yaml
index 0450ee429..a2ccc5e22 100644
--- a/providers/azure-open-ai/gpt-5-chat.yaml
+++ b/providers/azure-open-ai/gpt-5-chat.yaml
@@ -30,6 +30,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-mini-2025-08-07-lite.yaml b/providers/azure-open-ai/gpt-5-mini-2025-08-07-lite.yaml
index 9a56f63fb..6934edfcc 100644
--- a/providers/azure-open-ai/gpt-5-mini-2025-08-07-lite.yaml
+++ b/providers/azure-open-ai/gpt-5-mini-2025-08-07-lite.yaml
@@ -43,6 +43,7 @@ params:
       key: max_completion_tokens
       maxValue: 128000
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-mini-2025-08-07.yaml b/providers/azure-open-ai/gpt-5-mini-2025-08-07.yaml
index 6566abc00..9189cffe0 100644
--- a/providers/azure-open-ai/gpt-5-mini-2025-08-07.yaml
+++ b/providers/azure-open-ai/gpt-5-mini-2025-08-07.yaml
@@ -57,6 +57,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-mini.yaml b/providers/azure-open-ai/gpt-5-mini.yaml
index b28f36b3b..cadd1f6c9 100644
--- a/providers/azure-open-ai/gpt-5-mini.yaml
+++ b/providers/azure-open-ai/gpt-5-mini.yaml
@@ -56,6 +56,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5-nano-2025-08-07.yaml b/providers/azure-open-ai/gpt-5-nano-2025-08-07.yaml
index b7741cd80..42cfa850e 100644
--- a/providers/azure-open-ai/gpt-5-nano-2025-08-07.yaml
+++ b/providers/azure-open-ai/gpt-5-nano-2025-08-07.yaml
@@ -45,6 +45,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5-nano.yaml b/providers/azure-open-ai/gpt-5-nano.yaml
index bfb13136b..52ceb1614 100644
--- a/providers/azure-open-ai/gpt-5-nano.yaml
+++ b/providers/azure-open-ai/gpt-5-nano.yaml
@@ -58,6 +58,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.1-2025-11-13.yaml b/providers/azure-open-ai/gpt-5.1-2025-11-13.yaml
index 9809488b9..6d40aaf71 100644
--- a/providers/azure-open-ai/gpt-5.1-2025-11-13.yaml
+++ b/providers/azure-open-ai/gpt-5.1-2025-11-13.yaml
@@ -59,6 +59,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.1-chat-2025-11-13.yaml b/providers/azure-open-ai/gpt-5.1-chat-2025-11-13.yaml
index 322a017bc..916e69521 100644
--- a/providers/azure-open-ai/gpt-5.1-chat-2025-11-13.yaml
+++ b/providers/azure-open-ai/gpt-5.1-chat-2025-11-13.yaml
@@ -38,6 +38,7 @@ params:
     - defaultValue: null
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.1-chat.yaml b/providers/azure-open-ai/gpt-5.1-chat.yaml
index 6788b8f71..319517836 100644
--- a/providers/azure-open-ai/gpt-5.1-chat.yaml
+++ b/providers/azure-open-ai/gpt-5.1-chat.yaml
@@ -48,6 +48,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.1.yaml b/providers/azure-open-ai/gpt-5.1.yaml
index 6b80812b7..ca05a7620 100644
--- a/providers/azure-open-ai/gpt-5.1.yaml
+++ b/providers/azure-open-ai/gpt-5.1.yaml
@@ -59,6 +59,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.2-2025-12-11.yaml b/providers/azure-open-ai/gpt-5.2-2025-12-11.yaml
index ea6eabb27..110f9c135 100644
--- a/providers/azure-open-ai/gpt-5.2-2025-12-11.yaml
+++ b/providers/azure-open-ai/gpt-5.2-2025-12-11.yaml
@@ -43,6 +43,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.2-chat-2025-12-11.yaml b/providers/azure-open-ai/gpt-5.2-chat-2025-12-11.yaml
index c1b291a86..43a74c991 100644
--- a/providers/azure-open-ai/gpt-5.2-chat-2025-12-11.yaml
+++ b/providers/azure-open-ai/gpt-5.2-chat-2025-12-11.yaml
@@ -36,6 +36,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5.2-chat-2026-02-10.yaml b/providers/azure-open-ai/gpt-5.2-chat-2026-02-10.yaml
index edccaa931..22d327cc4 100644
--- a/providers/azure-open-ai/gpt-5.2-chat-2026-02-10.yaml
+++ b/providers/azure-open-ai/gpt-5.2-chat-2026-02-10.yaml
@@ -28,6 +28,7 @@ model: gpt-5.2-chat-2026-02-10
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5.2-chat.yaml b/providers/azure-open-ai/gpt-5.2-chat.yaml
index 1988d5376..37ba9490f 100644
--- a/providers/azure-open-ai/gpt-5.2-chat.yaml
+++ b/providers/azure-open-ai/gpt-5.2-chat.yaml
@@ -28,6 +28,7 @@ model: gpt-5.2-chat
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5.2.yaml b/providers/azure-open-ai/gpt-5.2.yaml
index 48df76716..57abeb815 100644
--- a/providers/azure-open-ai/gpt-5.2.yaml
+++ b/providers/azure-open-ai/gpt-5.2.yaml
@@ -47,6 +47,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-5.3-chat-2026-03-03.yaml b/providers/azure-open-ai/gpt-5.3-chat-2026-03-03.yaml
index 0567c99be..e540aa363 100644
--- a/providers/azure-open-ai/gpt-5.3-chat-2026-03-03.yaml
+++ b/providers/azure-open-ai/gpt-5.3-chat-2026-03-03.yaml
@@ -30,6 +30,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/gpt-5.yaml b/providers/azure-open-ai/gpt-5.yaml
index e5ead7b51..6f87f664e 100644
--- a/providers/azure-open-ai/gpt-5.yaml
+++ b/providers/azure-open-ai/gpt-5.yaml
@@ -55,6 +55,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/gpt-chat-latest-2026-05-05.yaml b/providers/azure-open-ai/gpt-chat-latest-2026-05-05.yaml
index 2dfe9d903..510aad1b4 100644
--- a/providers/azure-open-ai/gpt-chat-latest-2026-05-05.yaml
+++ b/providers/azure-open-ai/gpt-chat-latest-2026-05-05.yaml
@@ -23,6 +23,7 @@ modalities:
         - text
 mode: chat
 model: gpt-chat-latest-2026-05-05
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-08-05"
 sources:
diff --git a/providers/azure-open-ai/gpt-chat-latest-2026-05-28.yaml b/providers/azure-open-ai/gpt-chat-latest-2026-05-28.yaml
index 973cacddb..f6718b189 100644
--- a/providers/azure-open-ai/gpt-chat-latest-2026-05-28.yaml
+++ b/providers/azure-open-ai/gpt-chat-latest-2026-05-28.yaml
@@ -25,6 +25,7 @@ modalities:
         - text
 mode: chat
 model: gpt-chat-latest-2026-05-28
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-08-28"
 status: retired
diff --git a/providers/azure-open-ai/gpt-chat-latest-2026-06-24.yaml b/providers/azure-open-ai/gpt-chat-latest-2026-06-24.yaml
index 5f98309c6..106696300 100644
--- a/providers/azure-open-ai/gpt-chat-latest-2026-06-24.yaml
+++ b/providers/azure-open-ai/gpt-chat-latest-2026-06-24.yaml
@@ -24,6 +24,7 @@ modalities:
         - text
 mode: chat
 model: gpt-chat-latest-2026-06-24
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-09-24"
 status: preview
diff --git a/providers/azure-open-ai/gpt-chat-latest.yaml b/providers/azure-open-ai/gpt-chat-latest.yaml
index fa81ed236..6c774c832 100644
--- a/providers/azure-open-ai/gpt-chat-latest.yaml
+++ b/providers/azure-open-ai/gpt-chat-latest.yaml
@@ -30,6 +30,7 @@ mode: chat
 model: gpt-chat-latest
 params:
     - key: max_completion_tokens
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/model-router-2025-11-18.yaml b/providers/azure-open-ai/model-router-2025-11-18.yaml
index c52353f8e..9db0c296f 100644
--- a/providers/azure-open-ai/model-router-2025-11-18.yaml
+++ b/providers/azure-open-ai/model-router-2025-11-18.yaml
@@ -26,6 +26,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-05-20"
 sources:
diff --git a/providers/azure-open-ai/model-router.yaml b/providers/azure-open-ai/model-router.yaml
index a184a9aae..5a58a06a5 100644
--- a/providers/azure-open-ai/model-router.yaml
+++ b/providers/azure-open-ai/model-router.yaml
@@ -26,6 +26,7 @@ params:
     - key: max_completion_tokens
     - key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/o1-2024-12-17.yaml b/providers/azure-open-ai/o1-2024-12-17.yaml
index 04118d043..3ff0d5a05 100644
--- a/providers/azure-open-ai/o1-2024-12-17.yaml
+++ b/providers/azure-open-ai/o1-2024-12-17.yaml
@@ -19,6 +19,7 @@ modalities:
         - image
 mode: chat
 model: o1-2024-12-17
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/azure-open-ai/o1.yaml b/providers/azure-open-ai/o1.yaml
index 3c6a18fe2..e5f154298 100644
--- a/providers/azure-open-ai/o1.yaml
+++ b/providers/azure-open-ai/o1.yaml
@@ -29,6 +29,7 @@ params:
       key: max_completion_tokens
       maxValue: 100000
       minValue: 1
+preferChatCompletions: true
 removeParams:
     - temperature
     - top_p
diff --git a/providers/azure-open-ai/o3-2025-04-16.yaml b/providers/azure-open-ai/o3-2025-04-16.yaml
index 9be1a23f7..a8c33120c 100644
--- a/providers/azure-open-ai/o3-2025-04-16.yaml
+++ b/providers/azure-open-ai/o3-2025-04-16.yaml
@@ -36,6 +36,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/o3-deep-research-2025-06-26.yaml b/providers/azure-open-ai/o3-deep-research-2025-06-26.yaml
index 0118869a1..6056315ca 100644
--- a/providers/azure-open-ai/o3-deep-research-2025-06-26.yaml
+++ b/providers/azure-open-ai/o3-deep-research-2025-06-26.yaml
@@ -30,6 +30,7 @@ params:
       key: max_completion_tokens
       maxValue: 100000
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/o3-mini-2025-01-31.yaml b/providers/azure-open-ai/o3-mini-2025-01-31.yaml
index 204e1231a..4a74bf46b 100644
--- a/providers/azure-open-ai/o3-mini-2025-01-31.yaml
+++ b/providers/azure-open-ai/o3-mini-2025-01-31.yaml
@@ -40,6 +40,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/azure-open-ai/o3-mini.yaml b/providers/azure-open-ai/o3-mini.yaml
index 7e0230d3d..9c1dac923 100644
--- a/providers/azure-open-ai/o3-mini.yaml
+++ b/providers/azure-open-ai/o3-mini.yaml
@@ -50,6 +50,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/o3.yaml b/providers/azure-open-ai/o3.yaml
index 6190bc0cd..f88affc72 100644
--- a/providers/azure-open-ai/o3.yaml
+++ b/providers/azure-open-ai/o3.yaml
@@ -46,6 +46,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/o4-mini-2025-04-16.yaml b/providers/azure-open-ai/o4-mini-2025-04-16.yaml
index d8e0f75fd..eb5617662 100644
--- a/providers/azure-open-ai/o4-mini-2025-04-16.yaml
+++ b/providers/azure-open-ai/o4-mini-2025-04-16.yaml
@@ -35,6 +35,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/azure-open-ai/o4-mini.yaml b/providers/azure-open-ai/o4-mini.yaml
index 1a312ccc6..ca677a442 100644
--- a/providers/azure-open-ai/o4-mini.yaml
+++ b/providers/azure-open-ai/o4-mini.yaml
@@ -53,6 +53,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 removeParams:
     - top_p
     - max_tokens
diff --git a/providers/databricks/databricks-gpt-5-1.yaml b/providers/databricks/databricks-gpt-5-1.yaml
index af479ade3..fe38c8a18 100644
--- a/providers/databricks/databricks-gpt-5-1.yaml
+++ b/providers/databricks/databricks-gpt-5-1.yaml
@@ -42,6 +42,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/api-reference
diff --git a/providers/databricks/databricks-gpt-5-2.yaml b/providers/databricks/databricks-gpt-5-2.yaml
index 0e7013c9c..ef9ea0039 100644
--- a/providers/databricks/databricks-gpt-5-2.yaml
+++ b/providers/databricks/databricks-gpt-5-2.yaml
@@ -32,6 +32,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/supported-models
diff --git a/providers/databricks/databricks-gpt-5-3-codex.yaml b/providers/databricks/databricks-gpt-5-3-codex.yaml
index 81fd88faa..0971cbbd9 100644
--- a/providers/databricks/databricks-gpt-5-3-codex.yaml
+++ b/providers/databricks/databricks-gpt-5-3-codex.yaml
@@ -32,6 +32,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/model-serving/query-openai-responses
diff --git a/providers/databricks/databricks-gpt-5-mini.yaml b/providers/databricks/databricks-gpt-5-mini.yaml
index 88ac62a9d..debdfb596 100644
--- a/providers/databricks/databricks-gpt-5-mini.yaml
+++ b/providers/databricks/databricks-gpt-5-mini.yaml
@@ -34,6 +34,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/limits
diff --git a/providers/databricks/databricks-gpt-5-nano.yaml b/providers/databricks/databricks-gpt-5-nano.yaml
index e77afbc39..8755fd710 100644
--- a/providers/databricks/databricks-gpt-5-nano.yaml
+++ b/providers/databricks/databricks-gpt-5-nano.yaml
@@ -34,6 +34,7 @@ params:
           - xhigh
           - max
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://developers.openai.com/api/docs/models/gpt-5-nano
diff --git a/providers/databricks/databricks-gpt-5.yaml b/providers/databricks/databricks-gpt-5.yaml
index 41e596382..639b535d2 100644
--- a/providers/databricks/databricks-gpt-5.yaml
+++ b/providers/databricks/databricks-gpt-5.yaml
@@ -38,6 +38,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/api-reference
diff --git a/providers/databricks/system.ai.gpt-5-1.yaml b/providers/databricks/system.ai.gpt-5-1.yaml
index 2c49601ab..58a887c97 100644
--- a/providers/databricks/system.ai.gpt-5-1.yaml
+++ b/providers/databricks/system.ai.gpt-5-1.yaml
@@ -41,6 +41,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/api-reference
diff --git a/providers/databricks/system.ai.gpt-5-2.yaml b/providers/databricks/system.ai.gpt-5-2.yaml
index c56791306..29630642a 100644
--- a/providers/databricks/system.ai.gpt-5-2.yaml
+++ b/providers/databricks/system.ai.gpt-5-2.yaml
@@ -31,6 +31,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/model-serving/query-openai-responses
diff --git a/providers/databricks/system.ai.gpt-5-mini.yaml b/providers/databricks/system.ai.gpt-5-mini.yaml
index 6d9e5ac9f..f7252765e 100644
--- a/providers/databricks/system.ai.gpt-5-mini.yaml
+++ b/providers/databricks/system.ai.gpt-5-mini.yaml
@@ -35,6 +35,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/limits
diff --git a/providers/databricks/system.ai.gpt-5-nano.yaml b/providers/databricks/system.ai.gpt-5-nano.yaml
index 153255434..342a9b454 100644
--- a/providers/databricks/system.ai.gpt-5-nano.yaml
+++ b/providers/databricks/system.ai.gpt-5-nano.yaml
@@ -32,6 +32,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/model-serving/query-reason-models
diff --git a/providers/databricks/system.ai.gpt-5.yaml b/providers/databricks/system.ai.gpt-5.yaml
index 7a3b64bf1..42e39cc22 100644
--- a/providers/databricks/system.ai.gpt-5.yaml
+++ b/providers/databricks/system.ai.gpt-5.yaml
@@ -41,6 +41,7 @@ params:
           - xhigh
           - max
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/api-reference
diff --git a/providers/microsoft-foundry/computer-use-preview-2025-04-15.yaml b/providers/microsoft-foundry/computer-use-preview-2025-04-15.yaml
index a52354f3a..6f0a01a2f 100644
--- a/providers/microsoft-foundry/computer-use-preview-2025-04-15.yaml
+++ b/providers/microsoft-foundry/computer-use-preview-2025-04-15.yaml
@@ -27,6 +27,7 @@ params:
       key: max_tokens
       maxValue: 1024
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 status: deprecated
 supportedModes:
diff --git a/providers/microsoft-foundry/gpt-4.1-2025-04-14.yaml b/providers/microsoft-foundry/gpt-4.1-2025-04-14.yaml
index 5b7af641f..caca063e4 100644
--- a/providers/microsoft-foundry/gpt-4.1-2025-04-14.yaml
+++ b/providers/microsoft-foundry/gpt-4.1-2025-04-14.yaml
@@ -34,6 +34,7 @@ model: gpt-4.1-2025-04-14
 params:
     - key: max_tokens
       maxValue: 32768
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 sources:
diff --git a/providers/microsoft-foundry/gpt-4.1-mini-2025-04-14.yaml b/providers/microsoft-foundry/gpt-4.1-mini-2025-04-14.yaml
index 77a268e88..538dce061 100644
--- a/providers/microsoft-foundry/gpt-4.1-mini-2025-04-14.yaml
+++ b/providers/microsoft-foundry/gpt-4.1-mini-2025-04-14.yaml
@@ -36,6 +36,7 @@ model: gpt-4.1-mini-2025-04-14
 params:
     - key: max_tokens
       maxValue: 32768
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 sources:
diff --git a/providers/microsoft-foundry/gpt-4.1-mini.yaml b/providers/microsoft-foundry/gpt-4.1-mini.yaml
index 63b769cfe..9fb9a0e91 100644
--- a/providers/microsoft-foundry/gpt-4.1-mini.yaml
+++ b/providers/microsoft-foundry/gpt-4.1-mini.yaml
@@ -37,6 +37,7 @@ params:
       key: max_completion_tokens
       maxValue: 32768
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-4.1-nano-2025-04-14.yaml b/providers/microsoft-foundry/gpt-4.1-nano-2025-04-14.yaml
index 667f610f2..4ae7989fa 100644
--- a/providers/microsoft-foundry/gpt-4.1-nano-2025-04-14.yaml
+++ b/providers/microsoft-foundry/gpt-4.1-nano-2025-04-14.yaml
@@ -37,6 +37,7 @@ params:
       key: max_tokens
       maxValue: 32768
       minValue: 1
+preferChatCompletions: true
 removeParams:
     - reasoning_effort
 retirementDate: "2027-04-14"
diff --git a/providers/microsoft-foundry/gpt-4.1-nano.yaml b/providers/microsoft-foundry/gpt-4.1-nano.yaml
index d78c977cb..07a9de74e 100644
--- a/providers/microsoft-foundry/gpt-4.1-nano.yaml
+++ b/providers/microsoft-foundry/gpt-4.1-nano.yaml
@@ -40,6 +40,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-4.1.yaml b/providers/microsoft-foundry/gpt-4.1.yaml
index 09f898505..ec8c46f36 100644
--- a/providers/microsoft-foundry/gpt-4.1.yaml
+++ b/providers/microsoft-foundry/gpt-4.1.yaml
@@ -47,6 +47,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-4o-2024-05-13.yaml b/providers/microsoft-foundry/gpt-4o-2024-05-13.yaml
index da69ac3aa..0939eed7a 100644
--- a/providers/microsoft-foundry/gpt-4o-2024-05-13.yaml
+++ b/providers/microsoft-foundry/gpt-4o-2024-05-13.yaml
@@ -23,6 +23,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/microsoft-foundry/gpt-4o-2024-08-06.yaml b/providers/microsoft-foundry/gpt-4o-2024-08-06.yaml
index 5ba2bc10c..70ed82f11 100644
--- a/providers/microsoft-foundry/gpt-4o-2024-08-06.yaml
+++ b/providers/microsoft-foundry/gpt-4o-2024-08-06.yaml
@@ -33,6 +33,7 @@ params:
       key: max_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 sources:
     - https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/models
 status: deprecated
diff --git a/providers/microsoft-foundry/gpt-4o-2024-11-20.yaml b/providers/microsoft-foundry/gpt-4o-2024-11-20.yaml
index c62cb8272..4181063ea 100644
--- a/providers/microsoft-foundry/gpt-4o-2024-11-20.yaml
+++ b/providers/microsoft-foundry/gpt-4o-2024-11-20.yaml
@@ -39,6 +39,7 @@ model: gpt-4o-2024-11-20
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/microsoft-foundry/gpt-4o-mini-2024-07-18.yaml b/providers/microsoft-foundry/gpt-4o-mini-2024-07-18.yaml
index 36748a81e..43bc2ca54 100644
--- a/providers/microsoft-foundry/gpt-4o-mini-2024-07-18.yaml
+++ b/providers/microsoft-foundry/gpt-4o-mini-2024-07-18.yaml
@@ -38,6 +38,7 @@ model: gpt-4o-mini-2024-07-18
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 status: deprecated
diff --git a/providers/microsoft-foundry/gpt-4o-mini.yaml b/providers/microsoft-foundry/gpt-4o-mini.yaml
index 1464a6faa..b12e5ba13 100644
--- a/providers/microsoft-foundry/gpt-4o-mini.yaml
+++ b/providers/microsoft-foundry/gpt-4o-mini.yaml
@@ -40,6 +40,7 @@ params:
       key: max_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-04-14"
 sources:
diff --git a/providers/microsoft-foundry/gpt-4o.yaml b/providers/microsoft-foundry/gpt-4o.yaml
index d311471f9..26e0e5262 100644
--- a/providers/microsoft-foundry/gpt-4o.yaml
+++ b/providers/microsoft-foundry/gpt-4o.yaml
@@ -35,6 +35,7 @@ model: gpt-4o
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-10-01"
 status: deprecated
diff --git a/providers/microsoft-foundry/gpt-5-2025-08-07.yaml b/providers/microsoft-foundry/gpt-5-2025-08-07.yaml
index 696069a63..e587cbad9 100644
--- a/providers/microsoft-foundry/gpt-5-2025-08-07.yaml
+++ b/providers/microsoft-foundry/gpt-5-2025-08-07.yaml
@@ -43,6 +43,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-chat-2025-08-07.yaml b/providers/microsoft-foundry/gpt-5-chat-2025-08-07.yaml
index 3c1182347..ffd47c2b3 100644
--- a/providers/microsoft-foundry/gpt-5-chat-2025-08-07.yaml
+++ b/providers/microsoft-foundry/gpt-5-chat-2025-08-07.yaml
@@ -31,6 +31,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 removeParams:
     - max_tokens
 sources:
diff --git a/providers/microsoft-foundry/gpt-5-chat-2025-10-03.yaml b/providers/microsoft-foundry/gpt-5-chat-2025-10-03.yaml
index b69da4b74..127ba14b6 100644
--- a/providers/microsoft-foundry/gpt-5-chat-2025-10-03.yaml
+++ b/providers/microsoft-foundry/gpt-5-chat-2025-10-03.yaml
@@ -28,6 +28,7 @@ model: gpt-5-chat-2025-10-03
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-chat-latest.yaml b/providers/microsoft-foundry/gpt-5-chat-latest.yaml
index f9f390f41..30dcb6bae 100644
--- a/providers/microsoft-foundry/gpt-5-chat-latest.yaml
+++ b/providers/microsoft-foundry/gpt-5-chat-latest.yaml
@@ -29,6 +29,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-chat.yaml b/providers/microsoft-foundry/gpt-5-chat.yaml
index 0450ee429..a2ccc5e22 100644
--- a/providers/microsoft-foundry/gpt-5-chat.yaml
+++ b/providers/microsoft-foundry/gpt-5-chat.yaml
@@ -30,6 +30,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-mini-2025-08-07-lite.yaml b/providers/microsoft-foundry/gpt-5-mini-2025-08-07-lite.yaml
index e57bcbaa3..8d75a3cb5 100644
--- a/providers/microsoft-foundry/gpt-5-mini-2025-08-07-lite.yaml
+++ b/providers/microsoft-foundry/gpt-5-mini-2025-08-07-lite.yaml
@@ -49,6 +49,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-mini-2025-08-07.yaml b/providers/microsoft-foundry/gpt-5-mini-2025-08-07.yaml
index 6566abc00..9189cffe0 100644
--- a/providers/microsoft-foundry/gpt-5-mini-2025-08-07.yaml
+++ b/providers/microsoft-foundry/gpt-5-mini-2025-08-07.yaml
@@ -57,6 +57,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-mini.yaml b/providers/microsoft-foundry/gpt-5-mini.yaml
index 008c02194..9aea6e059 100644
--- a/providers/microsoft-foundry/gpt-5-mini.yaml
+++ b/providers/microsoft-foundry/gpt-5-mini.yaml
@@ -64,6 +64,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5-nano-2025-08-07.yaml b/providers/microsoft-foundry/gpt-5-nano-2025-08-07.yaml
index 344925c8d..102987d21 100644
--- a/providers/microsoft-foundry/gpt-5-nano-2025-08-07.yaml
+++ b/providers/microsoft-foundry/gpt-5-nano-2025-08-07.yaml
@@ -51,6 +51,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-5-nano.yaml b/providers/microsoft-foundry/gpt-5-nano.yaml
index bfb13136b..52ceb1614 100644
--- a/providers/microsoft-foundry/gpt-5-nano.yaml
+++ b/providers/microsoft-foundry/gpt-5-nano.yaml
@@ -58,6 +58,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.1-2025-11-13.yaml b/providers/microsoft-foundry/gpt-5.1-2025-11-13.yaml
index 5fbe7fd9b..cd1144cc8 100644
--- a/providers/microsoft-foundry/gpt-5.1-2025-11-13.yaml
+++ b/providers/microsoft-foundry/gpt-5.1-2025-11-13.yaml
@@ -57,6 +57,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.1-chat-2025-11-13.yaml b/providers/microsoft-foundry/gpt-5.1-chat-2025-11-13.yaml
index 5a1c1ee12..e4b2cc1d3 100644
--- a/providers/microsoft-foundry/gpt-5.1-chat-2025-11-13.yaml
+++ b/providers/microsoft-foundry/gpt-5.1-chat-2025-11-13.yaml
@@ -38,6 +38,7 @@ params:
     - defaultValue: null
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.1-chat.yaml b/providers/microsoft-foundry/gpt-5.1-chat.yaml
index 6788b8f71..319517836 100644
--- a/providers/microsoft-foundry/gpt-5.1-chat.yaml
+++ b/providers/microsoft-foundry/gpt-5.1-chat.yaml
@@ -48,6 +48,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.1.yaml b/providers/microsoft-foundry/gpt-5.1.yaml
index 6b80812b7..ca05a7620 100644
--- a/providers/microsoft-foundry/gpt-5.1.yaml
+++ b/providers/microsoft-foundry/gpt-5.1.yaml
@@ -59,6 +59,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.2-2025-12-11.yaml b/providers/microsoft-foundry/gpt-5.2-2025-12-11.yaml
index ea6eabb27..110f9c135 100644
--- a/providers/microsoft-foundry/gpt-5.2-2025-12-11.yaml
+++ b/providers/microsoft-foundry/gpt-5.2-2025-12-11.yaml
@@ -43,6 +43,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.2-chat-2025-12-11.yaml b/providers/microsoft-foundry/gpt-5.2-chat-2025-12-11.yaml
index 0c40fe3a7..08aadabb1 100644
--- a/providers/microsoft-foundry/gpt-5.2-chat-2025-12-11.yaml
+++ b/providers/microsoft-foundry/gpt-5.2-chat-2025-12-11.yaml
@@ -36,6 +36,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-05-13"
 sources:
diff --git a/providers/microsoft-foundry/gpt-5.2-chat-2026-02-10.yaml b/providers/microsoft-foundry/gpt-5.2-chat-2026-02-10.yaml
index 8a5719bdd..21f86c518 100644
--- a/providers/microsoft-foundry/gpt-5.2-chat-2026-02-10.yaml
+++ b/providers/microsoft-foundry/gpt-5.2-chat-2026-02-10.yaml
@@ -28,6 +28,7 @@ model: gpt-5.2-chat-2026-02-10
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-06-29"
 status: retired
diff --git a/providers/microsoft-foundry/gpt-5.2-chat.yaml b/providers/microsoft-foundry/gpt-5.2-chat.yaml
index e3a903617..91edbea26 100644
--- a/providers/microsoft-foundry/gpt-5.2-chat.yaml
+++ b/providers/microsoft-foundry/gpt-5.2-chat.yaml
@@ -28,6 +28,7 @@ model: gpt-5.2-chat
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-06-29"
 sources:
diff --git a/providers/microsoft-foundry/gpt-5.2.yaml b/providers/microsoft-foundry/gpt-5.2.yaml
index 48df76716..57abeb815 100644
--- a/providers/microsoft-foundry/gpt-5.2.yaml
+++ b/providers/microsoft-foundry/gpt-5.2.yaml
@@ -47,6 +47,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-5.3-chat-2026-03-03.yaml b/providers/microsoft-foundry/gpt-5.3-chat-2026-03-03.yaml
index fc46334ea..d6dc4e87c 100644
--- a/providers/microsoft-foundry/gpt-5.3-chat-2026-03-03.yaml
+++ b/providers/microsoft-foundry/gpt-5.3-chat-2026-03-03.yaml
@@ -30,6 +30,7 @@ params:
       key: max_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-06-29"
 sources:
diff --git a/providers/microsoft-foundry/gpt-5.yaml b/providers/microsoft-foundry/gpt-5.yaml
index 6061278eb..18c24a889 100644
--- a/providers/microsoft-foundry/gpt-5.yaml
+++ b/providers/microsoft-foundry/gpt-5.yaml
@@ -49,6 +49,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/gpt-chat-latest-2026-05-05.yaml b/providers/microsoft-foundry/gpt-chat-latest-2026-05-05.yaml
index 2dfe9d903..510aad1b4 100644
--- a/providers/microsoft-foundry/gpt-chat-latest-2026-05-05.yaml
+++ b/providers/microsoft-foundry/gpt-chat-latest-2026-05-05.yaml
@@ -23,6 +23,7 @@ modalities:
         - text
 mode: chat
 model: gpt-chat-latest-2026-05-05
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-08-05"
 sources:
diff --git a/providers/microsoft-foundry/gpt-chat-latest-2026-05-28.yaml b/providers/microsoft-foundry/gpt-chat-latest-2026-05-28.yaml
index a89632ed0..4dfb41f7c 100644
--- a/providers/microsoft-foundry/gpt-chat-latest-2026-05-28.yaml
+++ b/providers/microsoft-foundry/gpt-chat-latest-2026-05-28.yaml
@@ -28,6 +28,7 @@ model: gpt-chat-latest-2026-05-28
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-chat-latest-2026-06-24.yaml b/providers/microsoft-foundry/gpt-chat-latest-2026-06-24.yaml
index df82a731b..0b9abbff0 100644
--- a/providers/microsoft-foundry/gpt-chat-latest-2026-06-24.yaml
+++ b/providers/microsoft-foundry/gpt-chat-latest-2026-06-24.yaml
@@ -27,6 +27,7 @@ model: gpt-chat-latest-2026-06-24
 params:
     - key: max_completion_tokens
       maxValue: 16384
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-chat-latest-2026-08-06.yaml b/providers/microsoft-foundry/gpt-chat-latest-2026-08-06.yaml
index e4fb3043f..0d3718838 100644
--- a/providers/microsoft-foundry/gpt-chat-latest-2026-08-06.yaml
+++ b/providers/microsoft-foundry/gpt-chat-latest-2026-08-06.yaml
@@ -20,6 +20,7 @@ mode: chat
 model: gpt-chat-latest-2026-08-06
 params:
     - key: max_completion_tokens
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/gpt-chat-latest.yaml b/providers/microsoft-foundry/gpt-chat-latest.yaml
index e17dfe7ef..a1ca368e4 100644
--- a/providers/microsoft-foundry/gpt-chat-latest.yaml
+++ b/providers/microsoft-foundry/gpt-chat-latest.yaml
@@ -31,6 +31,7 @@ model: gpt-chat-latest
 params:
     - key: max_completion_tokens
       maxValue: 128000
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/grok-4.6.yaml b/providers/microsoft-foundry/grok-4.6.yaml
index 1e9098c9f..6eb0add4c 100644
--- a/providers/microsoft-foundry/grok-4.6.yaml
+++ b/providers/microsoft-foundry/grok-4.6.yaml
@@ -42,6 +42,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://techcommunity.microsoft.com/blog/azure-ai-foundry-blog/grok-4-6-comes-to-microsoft-foundry-models-built-for-long-horizon-reasoning-and-/4547578
diff --git a/providers/microsoft-foundry/model-router-2025-11-18.yaml b/providers/microsoft-foundry/model-router-2025-11-18.yaml
index c52353f8e..9db0c296f 100644
--- a/providers/microsoft-foundry/model-router-2025-11-18.yaml
+++ b/providers/microsoft-foundry/model-router-2025-11-18.yaml
@@ -26,6 +26,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-05-20"
 sources:
diff --git a/providers/microsoft-foundry/model-router.yaml b/providers/microsoft-foundry/model-router.yaml
index 0bd9dc50b..8b7cd286e 100644
--- a/providers/microsoft-foundry/model-router.yaml
+++ b/providers/microsoft-foundry/model-router.yaml
@@ -22,6 +22,7 @@ model: model-router
 params:
     - key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2027-05-20"
 sources:
diff --git a/providers/microsoft-foundry/o1-2024-12-17.yaml b/providers/microsoft-foundry/o1-2024-12-17.yaml
index 04118d043..3ff0d5a05 100644
--- a/providers/microsoft-foundry/o1-2024-12-17.yaml
+++ b/providers/microsoft-foundry/o1-2024-12-17.yaml
@@ -19,6 +19,7 @@ modalities:
         - image
 mode: chat
 model: o1-2024-12-17
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/microsoft-foundry/o1.yaml b/providers/microsoft-foundry/o1.yaml
index 3c6a18fe2..e5f154298 100644
--- a/providers/microsoft-foundry/o1.yaml
+++ b/providers/microsoft-foundry/o1.yaml
@@ -29,6 +29,7 @@ params:
       key: max_completion_tokens
       maxValue: 100000
       minValue: 1
+preferChatCompletions: true
 removeParams:
     - temperature
     - top_p
diff --git a/providers/microsoft-foundry/o3-2025-04-16.yaml b/providers/microsoft-foundry/o3-2025-04-16.yaml
index 8fe6ad6b1..d419d108b 100644
--- a/providers/microsoft-foundry/o3-2025-04-16.yaml
+++ b/providers/microsoft-foundry/o3-2025-04-16.yaml
@@ -39,6 +39,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/o3-deep-research-2025-06-26.yaml b/providers/microsoft-foundry/o3-deep-research-2025-06-26.yaml
index 0118869a1..6056315ca 100644
--- a/providers/microsoft-foundry/o3-deep-research-2025-06-26.yaml
+++ b/providers/microsoft-foundry/o3-deep-research-2025-06-26.yaml
@@ -30,6 +30,7 @@ params:
       key: max_completion_tokens
       maxValue: 100000
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/o3-mini-2025-01-31.yaml b/providers/microsoft-foundry/o3-mini-2025-01-31.yaml
index 204e1231a..4a74bf46b 100644
--- a/providers/microsoft-foundry/o3-mini-2025-01-31.yaml
+++ b/providers/microsoft-foundry/o3-mini-2025-01-31.yaml
@@ -40,6 +40,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/microsoft-foundry/o3-mini.yaml b/providers/microsoft-foundry/o3-mini.yaml
index 7e0230d3d..9c1dac923 100644
--- a/providers/microsoft-foundry/o3-mini.yaml
+++ b/providers/microsoft-foundry/o3-mini.yaml
@@ -50,6 +50,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/o3.yaml b/providers/microsoft-foundry/o3.yaml
index 6190bc0cd..f88affc72 100644
--- a/providers/microsoft-foundry/o3.yaml
+++ b/providers/microsoft-foundry/o3.yaml
@@ -46,6 +46,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/o4-mini-2025-04-16.yaml b/providers/microsoft-foundry/o4-mini-2025-04-16.yaml
index d8e0f75fd..eb5617662 100644
--- a/providers/microsoft-foundry/o4-mini-2025-04-16.yaml
+++ b/providers/microsoft-foundry/o4-mini-2025-04-16.yaml
@@ -35,6 +35,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/microsoft-foundry/o4-mini.yaml b/providers/microsoft-foundry/o4-mini.yaml
index 1a312ccc6..ca677a442 100644
--- a/providers/microsoft-foundry/o4-mini.yaml
+++ b/providers/microsoft-foundry/o4-mini.yaml
@@ -53,6 +53,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 removeParams:
     - top_p
     - max_tokens
diff --git a/providers/openai/chat-latest.yaml b/providers/openai/chat-latest.yaml
index a2cef6da1..3f04c648e 100644
--- a/providers/openai/chat-latest.yaml
+++ b/providers/openai/chat-latest.yaml
@@ -30,6 +30,7 @@ params:
       supportedValues:
           - medium
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/chatgpt-4o-latest.yaml b/providers/openai/chatgpt-4o-latest.yaml
index f8b5db3bb..1ed3ab9fa 100644
--- a/providers/openai/chatgpt-4o-latest.yaml
+++ b/providers/openai/chatgpt-4o-latest.yaml
@@ -19,6 +19,7 @@ modalities:
         - pdf
 mode: chat
 model: chatgpt-4o-latest
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/codex-mini-latest.yaml b/providers/openai/codex-mini-latest.yaml
index d87f3ed30..e428f4e73 100644
--- a/providers/openai/codex-mini-latest.yaml
+++ b/providers/openai/codex-mini-latest.yaml
@@ -6,5 +6,6 @@ costs:
 isDeprecated: true
 mode: chat
 model: codex-mini-latest
+preferChatCompletions: true
 supportedModes:
     - responses
diff --git a/providers/openai/gpt-3.5-turbo-0125.yaml b/providers/openai/gpt-3.5-turbo-0125.yaml
index ae5a96d0e..76248bb38 100644
--- a/providers/openai/gpt-3.5-turbo-0125.yaml
+++ b/providers/openai/gpt-3.5-turbo-0125.yaml
@@ -28,6 +28,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 sources:
     - https://developers.openai.com/api/docs/models/gpt-3.5-turbo
     - https://developers.openai.com/api/docs/guides/prompt-caching
diff --git a/providers/openai/gpt-3.5-turbo-0301.yaml b/providers/openai/gpt-3.5-turbo-0301.yaml
index a19e40be3..a45bc6adf 100644
--- a/providers/openai/gpt-3.5-turbo-0301.yaml
+++ b/providers/openai/gpt-3.5-turbo-0301.yaml
@@ -15,6 +15,7 @@ model: gpt-3.5-turbo-0301
 params:
     - key: max_tokens
       maxValue: 4096
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-3.5-turbo-0613.yaml b/providers/openai/gpt-3.5-turbo-0613.yaml
index 642086b9b..2452c5c7a 100644
--- a/providers/openai/gpt-3.5-turbo-0613.yaml
+++ b/providers/openai/gpt-3.5-turbo-0613.yaml
@@ -16,6 +16,7 @@ model: gpt-3.5-turbo-0613
 params:
     - key: max_tokens
       maxValue: 4096
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-3.5-turbo-1106.yaml b/providers/openai/gpt-3.5-turbo-1106.yaml
index 1746e08b5..d66b910d8 100644
--- a/providers/openai/gpt-3.5-turbo-1106.yaml
+++ b/providers/openai/gpt-3.5-turbo-1106.yaml
@@ -20,6 +20,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-3.5-turbo-16k-0613.yaml b/providers/openai/gpt-3.5-turbo-16k-0613.yaml
index 8945406c8..48e67cd1c 100644
--- a/providers/openai/gpt-3.5-turbo-16k-0613.yaml
+++ b/providers/openai/gpt-3.5-turbo-16k-0613.yaml
@@ -13,6 +13,7 @@ limits:
     max_tokens: 4096
 mode: chat
 model: gpt-3.5-turbo-16k-0613
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-3.5-turbo-16k.yaml b/providers/openai/gpt-3.5-turbo-16k.yaml
index 66993d192..fa97e38aa 100644
--- a/providers/openai/gpt-3.5-turbo-16k.yaml
+++ b/providers/openai/gpt-3.5-turbo-16k.yaml
@@ -16,6 +16,7 @@ model: gpt-3.5-turbo-16k
 params:
     - key: max_tokens
       maxValue: 16384
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-3.5-turbo-instruct.yaml b/providers/openai/gpt-3.5-turbo-instruct.yaml
index be1d4be3e..2dac9918f 100644
--- a/providers/openai/gpt-3.5-turbo-instruct.yaml
+++ b/providers/openai/gpt-3.5-turbo-instruct.yaml
@@ -17,6 +17,7 @@ modalities:
         - text
 mode: chat
 model: gpt-3.5-turbo-instruct
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-3.5-turbo.yaml b/providers/openai/gpt-3.5-turbo.yaml
index e5963a843..1addfc929 100644
--- a/providers/openai/gpt-3.5-turbo.yaml
+++ b/providers/openai/gpt-3.5-turbo.yaml
@@ -19,6 +19,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-4-0314.yaml b/providers/openai/gpt-4-0314.yaml
index 1bb5ad90d..57fb08a5e 100644
--- a/providers/openai/gpt-4-0314.yaml
+++ b/providers/openai/gpt-4-0314.yaml
@@ -15,6 +15,7 @@ model: gpt-4-0314
 params:
     - key: max_tokens
       maxValue: 8192
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-4-0613.yaml b/providers/openai/gpt-4-0613.yaml
index 40ff2bbc5..edfc8dc95 100644
--- a/providers/openai/gpt-4-0613.yaml
+++ b/providers/openai/gpt-4-0613.yaml
@@ -23,6 +23,7 @@ model: gpt-4-0613
 params:
     - key: max_tokens
       maxValue: 8192
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - tool_choice
diff --git a/providers/openai/gpt-4-32k-0314.yaml b/providers/openai/gpt-4-32k-0314.yaml
index 66fcd29c1..cba6ee5ae 100644
--- a/providers/openai/gpt-4-32k-0314.yaml
+++ b/providers/openai/gpt-4-32k-0314.yaml
@@ -15,6 +15,7 @@ model: gpt-4-32k-0314
 params:
     - key: max_tokens
       maxValue: 32768
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-4-32k.yaml b/providers/openai/gpt-4-32k.yaml
index f3541282e..3111a71f0 100644
--- a/providers/openai/gpt-4-32k.yaml
+++ b/providers/openai/gpt-4-32k.yaml
@@ -15,6 +15,7 @@ model: gpt-4-32k
 params:
     - key: max_tokens
       maxValue: 32768
+preferChatCompletions: true
 supportedModes:
     - chat
     - responses
diff --git a/providers/openai/gpt-4-turbo-2024-04-09.yaml b/providers/openai/gpt-4-turbo-2024-04-09.yaml
index d59840c13..744ce04ab 100644
--- a/providers/openai/gpt-4-turbo-2024-04-09.yaml
+++ b/providers/openai/gpt-4-turbo-2024-04-09.yaml
@@ -29,6 +29,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-10-23"
 sources:
diff --git a/providers/openai/gpt-4-turbo.yaml b/providers/openai/gpt-4-turbo.yaml
index ab6b95707..f94965455 100644
--- a/providers/openai/gpt-4-turbo.yaml
+++ b/providers/openai/gpt-4-turbo.yaml
@@ -30,6 +30,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 retirementDate: "2026-10-23"
 sources:
diff --git a/providers/openai/gpt-4.1-2025-04-14.yaml b/providers/openai/gpt-4.1-2025-04-14.yaml
index 12fdb11da..5fef45977 100644
--- a/providers/openai/gpt-4.1-2025-04-14.yaml
+++ b/providers/openai/gpt-4.1-2025-04-14.yaml
@@ -39,6 +39,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-4.1-mini-2025-04-14.yaml b/providers/openai/gpt-4.1-mini-2025-04-14.yaml
index 68c19617c..bb664df63 100644
--- a/providers/openai/gpt-4.1-mini-2025-04-14.yaml
+++ b/providers/openai/gpt-4.1-mini-2025-04-14.yaml
@@ -38,6 +38,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-4.1-mini.yaml b/providers/openai/gpt-4.1-mini.yaml
index b998a9b8d..87ea6ea4a 100644
--- a/providers/openai/gpt-4.1-mini.yaml
+++ b/providers/openai/gpt-4.1-mini.yaml
@@ -39,6 +39,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-4.1-nano-2025-04-14.yaml b/providers/openai/gpt-4.1-nano-2025-04-14.yaml
index ee24d873c..38f5df914 100644
--- a/providers/openai/gpt-4.1-nano-2025-04-14.yaml
+++ b/providers/openai/gpt-4.1-nano-2025-04-14.yaml
@@ -36,6 +36,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-4.1-nano.yaml b/providers/openai/gpt-4.1-nano.yaml
index 6f8097e8c..cce89cb50 100644
--- a/providers/openai/gpt-4.1-nano.yaml
+++ b/providers/openai/gpt-4.1-nano.yaml
@@ -34,6 +34,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-4.1.yaml b/providers/openai/gpt-4.1.yaml
index 72f1ff1f8..6e078b1d0 100644
--- a/providers/openai/gpt-4.1.yaml
+++ b/providers/openai/gpt-4.1.yaml
@@ -39,6 +39,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-4.yaml b/providers/openai/gpt-4.yaml
index c4cc3254d..e28ae6408 100644
--- a/providers/openai/gpt-4.yaml
+++ b/providers/openai/gpt-4.yaml
@@ -21,6 +21,7 @@ model: gpt-4
 params:
     - key: max_tokens
       maxValue: 8192
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - tool_choice
diff --git a/providers/openai/gpt-4o-2024-05-13.yaml b/providers/openai/gpt-4o-2024-05-13.yaml
index 8df21f405..cc7d8e74d 100644
--- a/providers/openai/gpt-4o-2024-05-13.yaml
+++ b/providers/openai/gpt-4o-2024-05-13.yaml
@@ -32,6 +32,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/openai/gpt-4o-2024-08-06.yaml b/providers/openai/gpt-4o-2024-08-06.yaml
index 80e8490d4..d4f1dead0 100644
--- a/providers/openai/gpt-4o-2024-08-06.yaml
+++ b/providers/openai/gpt-4o-2024-08-06.yaml
@@ -32,6 +32,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 sources:
     - https://developers.openai.com/api/docs/models/gpt-4o
diff --git a/providers/openai/gpt-4o-2024-11-20.yaml b/providers/openai/gpt-4o-2024-11-20.yaml
index 1dd10621e..de92d28b8 100644
--- a/providers/openai/gpt-4o-2024-11-20.yaml
+++ b/providers/openai/gpt-4o-2024-11-20.yaml
@@ -36,6 +36,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/openai/gpt-4o-mini-2024-07-18.yaml b/providers/openai/gpt-4o-mini-2024-07-18.yaml
index 4e1872436..31001581f 100644
--- a/providers/openai/gpt-4o-mini-2024-07-18.yaml
+++ b/providers/openai/gpt-4o-mini-2024-07-18.yaml
@@ -32,6 +32,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/openai/gpt-4o-mini.yaml b/providers/openai/gpt-4o-mini.yaml
index 3b93a3cfe..239595119 100644
--- a/providers/openai/gpt-4o-mini.yaml
+++ b/providers/openai/gpt-4o-mini.yaml
@@ -36,6 +36,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/openai/gpt-4o.yaml b/providers/openai/gpt-4o.yaml
index 95edb60fb..210db8f16 100644
--- a/providers/openai/gpt-4o.yaml
+++ b/providers/openai/gpt-4o.yaml
@@ -36,6 +36,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - reasoning_effort
diff --git a/providers/openai/gpt-5-2025-08-07.yaml b/providers/openai/gpt-5-2025-08-07.yaml
index f4a308c86..fdb50437d 100644
--- a/providers/openai/gpt-5-2025-08-07.yaml
+++ b/providers/openai/gpt-5-2025-08-07.yaml
@@ -36,6 +36,7 @@ params:
       type: string
     - defaultValue: null
       key: reasoning
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5-chat-latest.yaml b/providers/openai/gpt-5-chat-latest.yaml
index d5444cc8d..a1054eac1 100644
--- a/providers/openai/gpt-5-chat-latest.yaml
+++ b/providers/openai/gpt-5-chat-latest.yaml
@@ -35,6 +35,7 @@ params:
     - defaultValue: null
       key: response_format
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5-mini-2025-08-07.yaml b/providers/openai/gpt-5-mini-2025-08-07.yaml
index 6a4f38034..5b015dd01 100644
--- a/providers/openai/gpt-5-mini-2025-08-07.yaml
+++ b/providers/openai/gpt-5-mini-2025-08-07.yaml
@@ -38,6 +38,7 @@ params:
       key: verbosity
     - defaultValue: null
       key: reasoning
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5-mini.yaml b/providers/openai/gpt-5-mini.yaml
index f5a783eab..a00ead90c 100644
--- a/providers/openai/gpt-5-mini.yaml
+++ b/providers/openai/gpt-5-mini.yaml
@@ -52,6 +52,7 @@ params:
     - defaultValue: null
       key: verbosity
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5-nano-2025-08-07.yaml b/providers/openai/gpt-5-nano-2025-08-07.yaml
index b4d462de6..bf6eda637 100644
--- a/providers/openai/gpt-5-nano-2025-08-07.yaml
+++ b/providers/openai/gpt-5-nano-2025-08-07.yaml
@@ -38,6 +38,7 @@ params:
       key: verbosity
     - defaultValue: null
       key: reasoning
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5-nano.yaml b/providers/openai/gpt-5-nano.yaml
index 4788b638f..78cb370a1 100644
--- a/providers/openai/gpt-5-nano.yaml
+++ b/providers/openai/gpt-5-nano.yaml
@@ -37,6 +37,7 @@ params:
       key: verbosity
     - defaultValue: null
       key: reasoning
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5.1-2025-11-13.yaml b/providers/openai/gpt-5.1-2025-11-13.yaml
index e195555d2..2d79e4364 100644
--- a/providers/openai/gpt-5.1-2025-11-13.yaml
+++ b/providers/openai/gpt-5.1-2025-11-13.yaml
@@ -44,6 +44,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5.1-chat-latest.yaml b/providers/openai/gpt-5.1-chat-latest.yaml
index dd749562d..082379833 100644
--- a/providers/openai/gpt-5.1-chat-latest.yaml
+++ b/providers/openai/gpt-5.1-chat-latest.yaml
@@ -31,6 +31,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-5.1-codex-mini.yaml b/providers/openai/gpt-5.1-codex-mini.yaml
index afd100a95..79809c504 100644
--- a/providers/openai/gpt-5.1-codex-mini.yaml
+++ b/providers/openai/gpt-5.1-codex-mini.yaml
@@ -29,6 +29,7 @@ params:
       key: max_completion_tokens
       maxValue: 128000
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-5.1.yaml b/providers/openai/gpt-5.1.yaml
index ef227ae6a..6035e4fbd 100644
--- a/providers/openai/gpt-5.1.yaml
+++ b/providers/openai/gpt-5.1.yaml
@@ -44,6 +44,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - "n"
diff --git a/providers/openai/gpt-5.2-2025-12-11.yaml b/providers/openai/gpt-5.2-2025-12-11.yaml
index 318be0fd4..afafc818e 100644
--- a/providers/openai/gpt-5.2-2025-12-11.yaml
+++ b/providers/openai/gpt-5.2-2025-12-11.yaml
@@ -46,6 +46,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5.2-chat-latest.yaml b/providers/openai/gpt-5.2-chat-latest.yaml
index 2a430bba5..ea83e3a1c 100644
--- a/providers/openai/gpt-5.2-chat-latest.yaml
+++ b/providers/openai/gpt-5.2-chat-latest.yaml
@@ -39,6 +39,7 @@ params:
     - defaultValue: null
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/gpt-5.2-codex.yaml b/providers/openai/gpt-5.2-codex.yaml
index efd9e4be3..fcce45826 100644
--- a/providers/openai/gpt-5.2-codex.yaml
+++ b/providers/openai/gpt-5.2-codex.yaml
@@ -31,6 +31,7 @@ params:
       minValue: 1
     - defaultValue: high
       key: reasoning_effort
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-5.2.yaml b/providers/openai/gpt-5.2.yaml
index f75913418..a4512e897 100644
--- a/providers/openai/gpt-5.2.yaml
+++ b/providers/openai/gpt-5.2.yaml
@@ -45,6 +45,7 @@ params:
           - high
           - xhigh
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - "n"
diff --git a/providers/openai/gpt-5.3-chat-latest.yaml b/providers/openai/gpt-5.3-chat-latest.yaml
index edbf21fb0..b331a1fdf 100644
--- a/providers/openai/gpt-5.3-chat-latest.yaml
+++ b/providers/openai/gpt-5.3-chat-latest.yaml
@@ -30,6 +30,7 @@ params:
       key: max_completion_tokens
       maxValue: 16384
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/gpt-5.yaml b/providers/openai/gpt-5.yaml
index 2ff6f24fb..7be47f0a1 100644
--- a/providers/openai/gpt-5.yaml
+++ b/providers/openai/gpt-5.yaml
@@ -50,6 +50,7 @@ params:
           - medium
           - high
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/o1-2024-12-17.yaml b/providers/openai/o1-2024-12-17.yaml
index d74518463..2e37a9ed1 100644
--- a/providers/openai/o1-2024-12-17.yaml
+++ b/providers/openai/o1-2024-12-17.yaml
@@ -37,6 +37,7 @@ params:
       key: max_completion_tokens
       maxValue: 100000
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/o1.yaml b/providers/openai/o1.yaml
index dccc3637e..b01146e2d 100644
--- a/providers/openai/o1.yaml
+++ b/providers/openai/o1.yaml
@@ -36,6 +36,7 @@ params:
       key: max_completion_tokens
       maxValue: 100000
       minValue: 1
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/o3-2025-04-16.yaml b/providers/openai/o3-2025-04-16.yaml
index 39d5d2bec..6cf599287 100644
--- a/providers/openai/o3-2025-04-16.yaml
+++ b/providers/openai/o3-2025-04-16.yaml
@@ -37,6 +37,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/o3-mini-2025-01-31.yaml b/providers/openai/o3-mini-2025-01-31.yaml
index 5105314ba..ce3ac00f2 100644
--- a/providers/openai/o3-mini-2025-01-31.yaml
+++ b/providers/openai/o3-mini-2025-01-31.yaml
@@ -41,6 +41,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/o3-mini.yaml b/providers/openai/o3-mini.yaml
index 819d40fc7..741548984 100644
--- a/providers/openai/o3-mini.yaml
+++ b/providers/openai/o3-mini.yaml
@@ -41,6 +41,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - temperature
diff --git a/providers/openai/o3.yaml b/providers/openai/o3.yaml
index fdc08a746..610d8cd26 100644
--- a/providers/openai/o3.yaml
+++ b/providers/openai/o3.yaml
@@ -32,6 +32,7 @@ params:
       minValue: 1
     - defaultValue: medium
       key: reasoning_effort
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - max_tokens
diff --git a/providers/openai/o4-mini-2025-04-16.yaml b/providers/openai/o4-mini-2025-04-16.yaml
index 0da8c4f57..7787867c1 100644
--- a/providers/openai/o4-mini-2025-04-16.yaml
+++ b/providers/openai/o4-mini-2025-04-16.yaml
@@ -44,6 +44,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - top_p
diff --git a/providers/openai/o4-mini.yaml b/providers/openai/o4-mini.yaml
index 078238371..b193df5c1 100644
--- a/providers/openai/o4-mini.yaml
+++ b/providers/openai/o4-mini.yaml
@@ -43,6 +43,7 @@ params:
     - defaultValue: medium
       key: reasoning_effort
       type: string
+preferChatCompletions: true
 provisioning: serverless
 removeParams:
     - top_p