fix: align grok-4.5 and grok-4.6 params with xAI docs

models/xai/grok-4.5-subscription.yaml

@@ -40,10 +40,9 @@ params:
   - path: reasoning_effort
     type: enum
     label: Reasoning effort
-    description: Controls how much reasoning Grok performs before responding. Set to none for non-reasoning requests.
-    default: low
+    description: Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled.
+    default: high
     values:
-      - none
       - low
       - medium
       - high

models/xai/grok-4.5.yaml

@@ -40,10 +40,9 @@ params:
   - path: reasoning_effort
     type: enum
     label: Reasoning effort
-    description: Controls how much reasoning Grok performs before responding. Set to none for non-reasoning requests.
-    default: low
+    description: Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled.
+    default: high
     values:
-      - none
       - low
       - medium
       - high

models/xai/grok-4.6-subscription.yaml

@@ -0,0 +1,60 @@
+# yaml-language-server: $schema=https://modelparams.dev/api/v1/schema.json
+provider: xai
+authType: subscription
+apiSurface: openai-chat-completions
+model: grok-4.6
+status: active
+params:
+  - path: max_completion_tokens
+    type: integer
+    label: Max completion tokens
+    description: Upper bound for visible output tokens generated in the chat completion.
+    range:
+      min: 1
+    group: generation_length
+  - path: temperature
+    type: number
+    label: Temperature
+    description: Controls randomness. Lower values make outputs more focused; higher values make them more varied.
+    default: 1
+    range:
+      min: 0
+      max: 2
+      step: 0.1
+    group: sampling
+  - path: top_p
+    type: number
+    label: Top P
+    description: Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability.
+    default: 1
+    range:
+      min: 0
+      max: 1
+      step: 0.01
+    group: sampling
+  - path: seed
+    type: integer
+    label: Seed
+    description: Optional seed used for decoding when reproducible sampling is desired.
+    group: sampling
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled. xhigh is grok-4.6 and later.
+    default: high
+    values:
+      - low
+      - medium
+      - high
+      - xhigh
+    group: reasoning
+  - path: response_format.type
+    type: enum
+    label: Response format
+    description: Controls whether the model returns text, JSON mode output, or structured JSON schema output.
+    default: text
+    values:
+      - text
+      - json_object
+      - json_schema
+    group: output_format

models/xai/grok-4.6.yaml

@@ -3,6 +3,7 @@ provider: xai
 authType: api_key
 apiSurface: openai-chat-completions
 model: grok-4.6
+status: active
 params:
   - path: max_completion_tokens
     type: integer
@@ -39,8 +40,19 @@ params:
   - path: stop
     type: string
     label: Stop sequence
-    description: Stops generation when this sequence is produced. xAI accepts up to four stop sequences.
+    description: Stops generation when this sequence is produced. xAI accepts up to four stop sequences. grok-4.6 rejects stop because reasoning cannot be disabled.
     group: generation_length
+  - path: reasoning_effort
+    type: enum
+    label: Reasoning effort
+    description: Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled. xhigh is grok-4.6 and later.
+    default: high
+    values:
+      - low
+      - medium
+      - high
+      - xhigh
+    group: reasoning
   - path: response_format.type
     type: enum
     label: Response format

packages/modelparams-python/src/modelparams/_generated/catalog.json

@@ -27514,12 +27514,11 @@
       {
         "path": "reasoning_effort",
         "label": "Reasoning effort",
-        "description": "Controls how much reasoning Grok performs before responding. Set to none for non-reasoning requests.",
+        "description": "Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled.",
         "group": "reasoning",
         "type": "enum",
-        "default": "low",
+        "default": "high",
         "values": [
-          "none",
           "low",
           "medium",
           "high"
@@ -27593,12 +27592,11 @@
       {
         "path": "reasoning_effort",
         "label": "Reasoning effort",
-        "description": "Controls how much reasoning Grok performs before responding. Set to none for non-reasoning requests.",
+        "description": "Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled.",
         "group": "reasoning",
         "type": "enum",
-        "default": "low",
+        "default": "high",
         "values": [
-          "none",
           "low",
           "medium",
           "high"
@@ -27624,6 +27622,7 @@
     "authType": "api_key",
     "apiSurface": "openai-chat-completions",
     "model": "grok-4.6",
+    "status": "active",
     "params": [
       {
         "path": "max_completion_tokens",
@@ -27671,10 +27670,103 @@
       {
         "path": "stop",
         "label": "Stop sequence",
-        "description": "Stops generation when this sequence is produced. xAI accepts up to four stop sequences.",
+        "description": "Stops generation when this sequence is produced. xAI accepts up to four stop sequences. grok-4.6 rejects stop because reasoning cannot be disabled.",
         "group": "generation_length",
         "type": "string"
       },
+      {
+        "path": "reasoning_effort",
+        "label": "Reasoning effort",
+        "description": "Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled. xhigh is grok-4.6 and later.",
+        "group": "reasoning",
+        "type": "enum",
+        "default": "high",
+        "values": [
+          "low",
+          "medium",
+          "high",
+          "xhigh"
+        ]
+      },
+      {
+        "path": "response_format.type",
+        "label": "Response format",
+        "description": "Controls whether the model returns text, JSON mode output, or structured JSON schema output.",
+        "group": "output_format",
+        "type": "enum",
+        "default": "text",
+        "values": [
+          "text",
+          "json_object",
+          "json_schema"
+        ]
+      }
+    ]
+  },
+  {
+    "provider": "xai",
+    "authType": "subscription",
+    "apiSurface": "openai-chat-completions",
+    "model": "grok-4.6",
+    "status": "active",
+    "params": [
+      {
+        "path": "max_completion_tokens",
+        "label": "Max completion tokens",
+        "description": "Upper bound for visible output tokens generated in the chat completion.",
+        "group": "generation_length",
+        "type": "integer",
+        "range": {
+          "min": 1
+        }
+      },
+      {
+        "path": "temperature",
+        "label": "Temperature",
+        "description": "Controls randomness. Lower values make outputs more focused; higher values make them more varied.",
+        "group": "sampling",
+        "type": "number",
+        "default": 1,
+        "range": {
+          "min": 0,
+          "max": 2,
+          "step": 0.1
+        }
+      },
+      {
+        "path": "top_p",
+        "label": "Top P",
+        "description": "Controls nucleus sampling by limiting generation to tokens within the selected cumulative probability.",
+        "group": "sampling",
+        "type": "number",
+        "default": 1,
+        "range": {
+          "min": 0,
+          "max": 1,
+          "step": 0.01
+        }
+      },
+      {
+        "path": "seed",
+        "label": "Seed",
+        "description": "Optional seed used for decoding when reproducible sampling is desired.",
+        "group": "sampling",
+        "type": "integer"
+      },
+      {
+        "path": "reasoning_effort",
+        "label": "Reasoning effort",
+        "description": "Controls how much reasoning Grok performs before responding. Defaults to high. Reasoning cannot be disabled. xhigh is grok-4.6 and later.",
+        "group": "reasoning",
+        "type": "enum",
+        "default": "high",
+        "values": [
+          "low",
+          "medium",
+          "high",
+          "xhigh"
+        ]
+      },
       {
         "path": "response_format.type",
         "label": "Response format",

packages/modelparams-python/src/modelparams/_generated/model_ids.py

@@ -359,6 +359,7 @@
     "xai/grok-4.5",
     "xai/grok-4.5-subscription",
     "xai/grok-4.6",
+    "xai/grok-4.6-subscription",
     "xai/grok-build-0.1",
     "xai/grok-build-0.1-subscription",
     "xiaomi/mimo-v2.5",
@@ -744,6 +745,7 @@
     "xai/grok-4.5",
     "xai/grok-4.5-subscription",
     "xai/grok-4.6",
+    "xai/grok-4.6-subscription",
     "xai/grok-build-0.1",
     "xai/grok-build-0.1-subscription",
     "xiaomi/mimo-v2.5",

packages/modelparams-python/src/modelparams/_generated/registry.py

@@ -383,6 +383,7 @@
     "xai/grok-4.5": xai.Grok_4_5Params,
     "xai/grok-4.5-subscription": xai.Grok_4_5_SubscriptionParams,
     "xai/grok-4.6": xai.Grok_4_6Params,
+    "xai/grok-4.6-subscription": xai.Grok_4_6_SubscriptionParams,
     "xai/grok-build-0.1": xai.Grok_Build_0_1Params,
     "xai/grok-build-0.1-subscription": xai.Grok_Build_0_1_SubscriptionParams,
     "xiaomi/mimo-v2.5": xiaomi.Mimo_V2_5Params,

packages/modelparams-python/src/modelparams/types/xai.py

@@ -112,7 +112,7 @@
         "temperature": Annotated[float, Field(ge=0, le=2)],
         "top_p": Annotated[float, Field(ge=0, le=1)],
         "seed": int,
-        "reasoning_effort": Literal["none", "low", "medium", "high"],
+        "reasoning_effort": Literal["low", "medium", "high"],
         "response_format.type": Literal["text", "json_object", "json_schema"],
     },
     total=False,
@@ -126,7 +126,7 @@
         "temperature": Annotated[float, Field(ge=0, le=2)],
         "top_p": Annotated[float, Field(ge=0, le=1)],
         "seed": int,
-        "reasoning_effort": Literal["none", "low", "medium", "high"],
+        "reasoning_effort": Literal["low", "medium", "high"],
         "response_format.type": Literal["text", "json_object", "json_schema"],
     },
     total=False,
@@ -141,12 +141,27 @@
         "top_p": Annotated[float, Field(ge=0, le=1)],
         "seed": int,
         "stop": str,
+        "reasoning_effort": Literal["low", "medium", "high", "xhigh"],
         "response_format.type": Literal["text", "json_object", "json_schema"],
     },
     total=False,
 )
 setattr(Grok_4_6Params, "__pydantic_config__", _PARAMS_CONFIG)
 
+Grok_4_6_SubscriptionParams = TypedDict(
+    "Grok_4_6_SubscriptionParams",
+    {
+        "max_completion_tokens": Annotated[int, Field(ge=1)],
+        "temperature": Annotated[float, Field(ge=0, le=2)],
+        "top_p": Annotated[float, Field(ge=0, le=1)],
+        "seed": int,
+        "reasoning_effort": Literal["low", "medium", "high", "xhigh"],
+        "response_format.type": Literal["text", "json_object", "json_schema"],
+    },
+    total=False,
+)
+setattr(Grok_4_6_SubscriptionParams, "__pydantic_config__", _PARAMS_CONFIG)
+
 Grok_Build_0_1Params = TypedDict(
     "Grok_Build_0_1Params",
     {
@@ -184,6 +199,7 @@
     "Grok_4_5Params",
     "Grok_4_5_SubscriptionParams",
     "Grok_4_6Params",
+    "Grok_4_6_SubscriptionParams",
     "Grok_Build_0_1Params",
     "Grok_Build_0_1_SubscriptionParams",
 ]