diff --git a/models/z-ai/glm-4.5-air-subscription.yaml b/models/z-ai/glm-4.5-air-subscription.yaml index 9fba0d00..ca54a471 100644 --- a/models/z-ai/glm-4.5-air-subscription.yaml +++ b/models/z-ai/glm-4.5-air-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.5-air.yaml b/models/z-ai/glm-4.5-air.yaml index 68e43985..30e9a6fb 100644 --- a/models/z-ai/glm-4.5-air.yaml +++ b/models/z-ai/glm-4.5-air.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.5-airx.yaml b/models/z-ai/glm-4.5-airx.yaml index b000b314..b5136969 100644 --- a/models/z-ai/glm-4.5-airx.yaml +++ b/models/z-ai/glm-4.5-airx.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.5-flash.yaml b/models/z-ai/glm-4.5-flash.yaml index adf433e0..2ebfbc21 100644 --- a/models/z-ai/glm-4.5-flash.yaml +++ b/models/z-ai/glm-4.5-flash.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.5-subscription.yaml b/models/z-ai/glm-4.5-subscription.yaml index be775403..0de83a86 100644 --- a/models/z-ai/glm-4.5-subscription.yaml +++ b/models/z-ai/glm-4.5-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.5-x.yaml b/models/z-ai/glm-4.5-x.yaml index 027b3340..08c22ea6 100644 --- a/models/z-ai/glm-4.5-x.yaml +++ b/models/z-ai/glm-4.5-x.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.5.yaml b/models/z-ai/glm-4.5.yaml index b28d0ddb..d4f0ebee 100644 --- a/models/z-ai/glm-4.5.yaml +++ b/models/z-ai/glm-4.5.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 98304 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.6-subscription.yaml b/models/z-ai/glm-4.6-subscription.yaml index 933ef87f..3f7b20a3 100644 --- a/models/z-ai/glm-4.6-subscription.yaml +++ b/models/z-ai/glm-4.6-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.6.yaml b/models/z-ai/glm-4.6.yaml index 3301f344..3bba63c9 100644 --- a/models/z-ai/glm-4.6.yaml +++ b/models/z-ai/glm-4.6.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.7-flash.yaml b/models/z-ai/glm-4.7-flash.yaml index f558b359..c8320f8a 100644 --- a/models/z-ai/glm-4.7-flash.yaml +++ b/models/z-ai/glm-4.7-flash.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.7-flashx.yaml b/models/z-ai/glm-4.7-flashx.yaml index a53a1021..5d2f75e7 100644 --- a/models/z-ai/glm-4.7-flashx.yaml +++ b/models/z-ai/glm-4.7-flashx.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.7-subscription.yaml b/models/z-ai/glm-4.7-subscription.yaml index 79bbdc8c..fb225543 100644 --- a/models/z-ai/glm-4.7-subscription.yaml +++ b/models/z-ai/glm-4.7-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-4.7.yaml b/models/z-ai/glm-4.7.yaml index 889cb7a0..233f6b48 100644 --- a/models/z-ai/glm-4.7.yaml +++ b/models/z-ai/glm-4.7.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-5-subscription.yaml b/models/z-ai/glm-5-subscription.yaml index 53d60d3e..3a4e2aa2 100644 --- a/models/z-ai/glm-5-subscription.yaml +++ b/models/z-ai/glm-5-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-5-turbo-subscription.yaml b/models/z-ai/glm-5-turbo-subscription.yaml index 3045f096..67c35c88 100644 --- a/models/z-ai/glm-5-turbo-subscription.yaml +++ b/models/z-ai/glm-5-turbo-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-5-turbo.yaml b/models/z-ai/glm-5-turbo.yaml index fafce34f..0b6bfd37 100644 --- a/models/z-ai/glm-5-turbo.yaml +++ b/models/z-ai/glm-5-turbo.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-5.1-subscription.yaml b/models/z-ai/glm-5.1-subscription.yaml index b0eb9423..d6729d0f 100644 --- a/models/z-ai/glm-5.1-subscription.yaml +++ b/models/z-ai/glm-5.1-subscription.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-5.1.yaml b/models/z-ai/glm-5.1.yaml index bba79e5c..c3205033 100644 --- a/models/z-ai/glm-5.1.yaml +++ b/models/z-ai/glm-5.1.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/models/z-ai/glm-5.3.yaml b/models/z-ai/glm-5.3.yaml index 9488bae8..fc4fce5e 100644 --- a/models/z-ai/glm-5.3.yaml +++ b/models/z-ai/glm-5.3.yaml @@ -47,29 +47,21 @@ params: - path: thinking.type type: enum label: Thinking mode - description: Toggles the model's extended reasoning before it produces the final answer. + description: GLM-5.3 always engages in extended reasoning; thinking cannot be disabled. default: enabled values: - enabled - - disabled group: reasoning - path: reasoning_effort type: enum label: Reasoning effort - description: Controls how much reasoning effort GLM-5.2 spends when thinking is enabled. + description: Controls how much reasoning effort GLM-5.3 spends on its always-on thinking. default: max values: - - none - - minimal - low - - medium - high - - xhigh - max group: reasoning - applicability: - only: - thinking.type: enabled - path: response_format.type type: enum label: Response format diff --git a/models/z-ai/glm-5.yaml b/models/z-ai/glm-5.yaml index 82ab4c68..28d6944f 100644 --- a/models/z-ai/glm-5.yaml +++ b/models/z-ai/glm-5.yaml @@ -8,8 +8,10 @@ params: type: integer label: Max tokens description: Maximum number of tokens to generate in the response. + default: 65536 range: min: 1 + max: 131072 group: generation_length - path: temperature type: number diff --git a/packages/modelparams-python/src/modelparams/_generated/catalog.json b/packages/modelparams-python/src/modelparams/_generated/catalog.json index b092437f..53a055ab 100644 --- a/packages/modelparams-python/src/modelparams/_generated/catalog.json +++ b/packages/modelparams-python/src/modelparams/_generated/catalog.json @@ -27742,8 +27742,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -27828,8 +27830,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -27914,8 +27918,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28000,8 +28006,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28086,8 +28094,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28172,8 +28182,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28258,8 +28270,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28344,8 +28358,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28430,8 +28446,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28516,8 +28534,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28602,8 +28622,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28688,8 +28710,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28774,8 +28798,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28860,8 +28886,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28946,8 +28974,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29032,8 +29062,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29118,8 +29150,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29204,8 +29238,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29290,8 +29326,10 @@ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29648,34 +29686,24 @@ { "path": "thinking.type", "label": "Thinking mode", - "description": "Toggles the model's extended reasoning before it produces the final answer.", + "description": "GLM-5.3 always engages in extended reasoning; thinking cannot be disabled.", "group": "reasoning", "type": "enum", "default": "enabled", "values": [ - "enabled", - "disabled" + "enabled" ] }, { "path": "reasoning_effort", "label": "Reasoning effort", - "description": "Controls how much reasoning effort GLM-5.2 spends when thinking is enabled.", + "description": "Controls how much reasoning effort GLM-5.3 spends on its always-on thinking.", "group": "reasoning", - "applicability": { - "only": { - "thinking.type": "enabled" - } - }, "type": "enum", "default": "max", "values": [ - "none", - "minimal", "low", - "medium", "high", - "xhigh", "max" ] }, diff --git a/packages/modelparams-python/src/modelparams/types/z_ai.py b/packages/modelparams-python/src/modelparams/types/z_ai.py index 46b9c552..6a84c4c8 100644 --- a/packages/modelparams-python/src/modelparams/types/z_ai.py +++ b/packages/modelparams-python/src/modelparams/types/z_ai.py @@ -13,7 +13,7 @@ Glm_4_5Params = TypedDict( "Glm_4_5Params", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -27,7 +27,7 @@ Glm_4_5_AirParams = TypedDict( "Glm_4_5_AirParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -41,7 +41,7 @@ Glm_4_5_Air_SubscriptionParams = TypedDict( "Glm_4_5_Air_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -55,7 +55,7 @@ Glm_4_5_AirxParams = TypedDict( "Glm_4_5_AirxParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -69,7 +69,7 @@ Glm_4_5_FlashParams = TypedDict( "Glm_4_5_FlashParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -83,7 +83,7 @@ Glm_4_5_SubscriptionParams = TypedDict( "Glm_4_5_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -97,7 +97,7 @@ Glm_4_5_XParams = TypedDict( "Glm_4_5_XParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=98304)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -111,7 +111,7 @@ Glm_4_6Params = TypedDict( "Glm_4_6Params", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -125,7 +125,7 @@ Glm_4_6_SubscriptionParams = TypedDict( "Glm_4_6_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -139,7 +139,7 @@ Glm_4_7Params = TypedDict( "Glm_4_7Params", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -153,7 +153,7 @@ Glm_4_7_FlashParams = TypedDict( "Glm_4_7_FlashParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -167,7 +167,7 @@ Glm_4_7_FlashxParams = TypedDict( "Glm_4_7_FlashxParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -181,7 +181,7 @@ Glm_4_7_SubscriptionParams = TypedDict( "Glm_4_7_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -195,7 +195,7 @@ Glm_5Params = TypedDict( "Glm_5Params", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -209,7 +209,7 @@ Glm_5_SubscriptionParams = TypedDict( "Glm_5_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -223,7 +223,7 @@ Glm_5_TurboParams = TypedDict( "Glm_5_TurboParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -237,7 +237,7 @@ Glm_5_Turbo_SubscriptionParams = TypedDict( "Glm_5_Turbo_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -251,7 +251,7 @@ Glm_5_1Params = TypedDict( "Glm_5_1Params", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -265,7 +265,7 @@ Glm_5_1_SubscriptionParams = TypedDict( "Glm_5_1_SubscriptionParams", { - "max_tokens": Annotated[int, Field(ge=1)], + "max_tokens": Annotated[int, Field(ge=1, le=131072)], "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, @@ -313,8 +313,8 @@ "temperature": Annotated[float, Field(ge=0, le=1)], "top_p": Annotated[float, Field(ge=0.01, le=1)], "do_sample": bool, - "thinking.type": Literal["enabled", "disabled"], - "reasoning_effort": Literal["none", "minimal", "low", "medium", "high", "xhigh", "max"], + "thinking.type": Literal["enabled"], + "reasoning_effort": Literal["low", "high", "max"], "response_format.type": Literal["text", "json_object"], }, total=False, diff --git a/packages/modelparams/src/generated/data.ts b/packages/modelparams/src/generated/data.ts index cbc805e5..e5832936 100644 --- a/packages/modelparams/src/generated/data.ts +++ b/packages/modelparams/src/generated/data.ts @@ -27747,8 +27747,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -27833,8 +27835,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -27919,8 +27923,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28005,8 +28011,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28091,8 +28099,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28177,8 +28187,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28263,8 +28275,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 98304 } }, { @@ -28349,8 +28363,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28435,8 +28451,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28521,8 +28539,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28607,8 +28627,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28693,8 +28715,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28779,8 +28803,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28865,8 +28891,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -28951,8 +28979,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29037,8 +29067,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29123,8 +29155,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29209,8 +29243,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29295,8 +29331,10 @@ const GENERATED_CATALOG = [ "description": "Maximum number of tokens to generate in the response.", "group": "generation_length", "type": "integer", + "default": 65536, "range": { - "min": 1 + "min": 1, + "max": 131072 } }, { @@ -29653,34 +29691,24 @@ const GENERATED_CATALOG = [ { "path": "thinking.type", "label": "Thinking mode", - "description": "Toggles the model's extended reasoning before it produces the final answer.", + "description": "GLM-5.3 always engages in extended reasoning; thinking cannot be disabled.", "group": "reasoning", "type": "enum", "default": "enabled", "values": [ - "enabled", - "disabled" + "enabled" ] }, { "path": "reasoning_effort", "label": "Reasoning effort", - "description": "Controls how much reasoning effort GLM-5.2 spends when thinking is enabled.", + "description": "Controls how much reasoning effort GLM-5.3 spends on its always-on thinking.", "group": "reasoning", - "applicability": { - "only": { - "thinking.type": "enabled" - } - }, "type": "enum", "default": "max", "values": [ - "none", - "minimal", "low", - "medium", "high", - "xhigh", "max" ] }, diff --git a/packages/modelparams/src/generated/defaults.ts b/packages/modelparams/src/generated/defaults.ts index ab9b1bea..18b18734 100644 --- a/packages/modelparams/src/generated/defaults.ts +++ b/packages/modelparams/src/generated/defaults.ts @@ -1816,6 +1816,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1823,6 +1824,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5-air": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1830,6 +1832,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5-air-subscription": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1837,6 +1840,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5-airx": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1844,6 +1848,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5-flash": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1851,6 +1856,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5-subscription": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1858,6 +1864,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.5-x": { + max_tokens: 65536, temperature: 0.6, top_p: 0.95, do_sample: true, @@ -1865,6 +1872,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.6": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1872,6 +1880,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.6-subscription": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1879,6 +1888,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.7": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1886,6 +1896,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.7-flash": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1893,6 +1904,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.7-flashx": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1900,6 +1912,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-4.7-subscription": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1907,6 +1920,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-5": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1914,6 +1928,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-5-subscription": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1921,6 +1936,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-5-turbo": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1928,6 +1944,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-5-turbo-subscription": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1935,6 +1952,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-5.1": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, @@ -1942,6 +1960,7 @@ export const DEFAULTS = { "response_format.type": "text", }, "z-ai/glm-5.1-subscription": { + max_tokens: 65536, temperature: 1, top_p: 0.95, do_sample: true, diff --git a/packages/modelparams/src/generated/params-by-id.ts b/packages/modelparams/src/generated/params-by-id.ts index 74a07bb4..d4685893 100644 --- a/packages/modelparams/src/generated/params-by-id.ts +++ b/packages/modelparams/src/generated/params-by-id.ts @@ -2988,8 +2988,8 @@ export type ParamsById = { temperature: number; top_p: number; do_sample: boolean; - "thinking.type": "enabled" | "disabled"; - reasoning_effort: "none" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max"; + "thinking.type": "enabled"; + reasoning_effort: "low" | "high" | "max"; "response_format.type": "text" | "json_object"; }; };