modelparams 0.0.13__tar.gz → 0.0.15__tar.gz

{modelparams-0.0.13 → modelparams-0.0.15}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: modelparams
-Version: 0.0.13
+Version: 0.0.15
 Summary: Typed model parameters for Python, generated from the modelparams.dev catalog.
 Project-URL: Homepage, https://modelparams.dev
 Project-URL: Repository, https://github.com/mnfst/modelparameters.dev

{modelparams-0.0.13 → modelparams-0.0.15}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "modelparams"
-version = "0.0.13"
+version = "0.0.15"
 description = "Typed model parameters for Python, generated from the modelparams.dev catalog."
 readme = "README.md"
 requires-python = ">=3.10"

{modelparams-0.0.13 → modelparams-0.0.15}/src/modelparams/_generated/catalog.json

@@ -59,6 +59,66 @@
       }
     ]
   },
+  {
+    "provider": "alibaba",
+    "authType": "api_key",
+    "model": "qwen-max",
+    "params": [
+      {
+        "path": "max_tokens",
+        "label": "Max tokens",
+        "description": "Maximum number of output tokens the model may generate.",
+        "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",
+        "range": {
+          "min": 0,
+          "max": 1.9,
+          "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",
+        "range": {
+          "min": 0,
+          "max": 1,
+          "step": 0.01
+        }
+      },
+      {
+        "path": "extra_body.top_k",
+        "label": "Top K",
+        "description": "Limits generation to the selected number of highest-probability tokens.",
+        "group": "sampling",
+        "type": "integer",
+        "default": 20,
+        "range": {
+          "min": 1
+        }
+      },
+      {
+        "path": "extra_body.chat_template_kwargs.enable_thinking",
+        "label": "Enable thinking",
+        "description": "Controls Qwen3 thinking mode when using OpenAI-compatible clients that pass provider-specific extra body fields.",
+        "group": "reasoning",
+        "type": "boolean",
+        "default": true
+      }
+    ]
+  },
   {
     "provider": "alibaba",
     "authType": "api_key",
@@ -119,6 +179,66 @@
       }
     ]
   },
+  {
+    "provider": "alibaba",
+    "authType": "api_key",
+    "model": "qwen-turbo",
+    "params": [
+      {
+        "path": "max_tokens",
+        "label": "Max tokens",
+        "description": "Maximum number of output tokens the model may generate.",
+        "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",
+        "range": {
+          "min": 0,
+          "max": 1.9,
+          "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",
+        "range": {
+          "min": 0,
+          "max": 1,
+          "step": 0.01
+        }
+      },
+      {
+        "path": "extra_body.top_k",
+        "label": "Top K",
+        "description": "Limits generation to the selected number of highest-probability tokens.",
+        "group": "sampling",
+        "type": "integer",
+        "default": 20,
+        "range": {
+          "min": 1
+        }
+      },
+      {
+        "path": "extra_body.chat_template_kwargs.enable_thinking",
+        "label": "Enable thinking",
+        "description": "Controls Qwen3 thinking mode when using OpenAI-compatible clients that pass provider-specific extra body fields.",
+        "group": "reasoning",
+        "type": "boolean",
+        "default": true
+      }
+    ]
+  },
   {
     "provider": "alibaba",
     "authType": "api_key",
@@ -628,6 +748,81 @@
       }
     ]
   },
+  {
+    "provider": "alibaba",
+    "authType": "api_key",
+    "model": "qwen3.8-2.4t-a95b",
+    "params": [
+      {
+        "path": "max_completion_tokens",
+        "label": "Max tokens",
+        "description": "Maximum number of tokens to generate, including both reasoning and the final answer.",
+        "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",
+        "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",
+        "range": {
+          "min": 0,
+          "max": 1,
+          "step": 0.01
+        }
+      },
+      {
+        "path": "extra_body.top_k",
+        "label": "Top K",
+        "description": "Limits generation to the selected number of highest-probability tokens. Values above 100 disable top-k sampling.",
+        "group": "sampling",
+        "type": "integer",
+        "default": 20,
+        "range": {
+          "min": 0
+        }
+      },
+      {
+        "path": "extra_body.enable_thinking",
+        "label": "Enable thinking",
+        "description": "Toggles the model's hybrid thinking mode, sent as a provider-specific extra body field on OpenAI-compatible clients.",
+        "group": "reasoning",
+        "type": "boolean",
+        "default": true
+      },
+      {
+        "path": "extra_body.thinking_budget",
+        "label": "Thinking budget",
+        "description": "Maximum number of tokens the model may spend on reasoning before it starts the final answer; defaults to the model's maximum reasoning length.",
+        "group": "reasoning",
+        "applicability": {
+          "only": {
+            "extra_body.enable_thinking": true
+          }
+        },
+        "type": "integer",
+        "range": {
+          "min": 1
+        }
+      }
+    ]
+  },
   {
     "provider": "alibaba",
     "authType": "api_key",

{modelparams-0.0.13 → modelparams-0.0.15}/src/modelparams/_generated/model_ids.py

@@ -5,7 +5,9 @@ from typing import Literal
 
 ModelId = Literal[
     "alibaba/qwen-flash",
+    "alibaba/qwen-max",
     "alibaba/qwen-plus",
+    "alibaba/qwen-turbo",
     "alibaba/qwen3-coder-flash",
     "alibaba/qwen3-coder-plus",
     "alibaba/qwen3-max",
@@ -14,6 +16,7 @@ ModelId = Literal[
     "alibaba/qwen3.6-flash",
     "alibaba/qwen3.7-max",
     "alibaba/qwen3.7-plus",
+    "alibaba/qwen3.8-2.4t-a95b",
     "alibaba/qwen3.8-max",
     "alibaba/qwq-plus",
     "anthropic/claude-3-5-haiku-20241022",
@@ -263,7 +266,9 @@ ModelId = Literal[
 
 MODEL_IDS: tuple[ModelId, ...] = (
     "alibaba/qwen-flash",
+    "alibaba/qwen-max",
     "alibaba/qwen-plus",
+    "alibaba/qwen-turbo",
     "alibaba/qwen3-coder-flash",
     "alibaba/qwen3-coder-plus",
     "alibaba/qwen3-max",
@@ -272,6 +277,7 @@ MODEL_IDS: tuple[ModelId, ...] = (
     "alibaba/qwen3.6-flash",
     "alibaba/qwen3.7-max",
     "alibaba/qwen3.7-plus",
+    "alibaba/qwen3.8-2.4t-a95b",
     "alibaba/qwen3.8-max",
     "alibaba/qwq-plus",
     "anthropic/claude-3-5-haiku-20241022",

{modelparams-0.0.13 → modelparams-0.0.15}/src/modelparams/_generated/registry.py

@@ -26,7 +26,9 @@ from .model_ids import ModelId
 
 PARAM_TYPES: dict[ModelId, Any] = {
     "alibaba/qwen-flash": alibaba.Qwen_FlashParams,
+    "alibaba/qwen-max": alibaba.Qwen_MaxParams,
     "alibaba/qwen-plus": alibaba.Qwen_PlusParams,
+    "alibaba/qwen-turbo": alibaba.Qwen_TurboParams,
     "alibaba/qwen3-coder-flash": alibaba.Qwen3_Coder_FlashParams,
     "alibaba/qwen3-coder-plus": alibaba.Qwen3_Coder_PlusParams,
     "alibaba/qwen3-max": alibaba.Qwen3_MaxParams,
@@ -35,6 +37,7 @@ PARAM_TYPES: dict[ModelId, Any] = {
     "alibaba/qwen3.6-flash": alibaba.Qwen3_6_FlashParams,
     "alibaba/qwen3.7-max": alibaba.Qwen3_7_MaxParams,
     "alibaba/qwen3.7-plus": alibaba.Qwen3_7_PlusParams,
+    "alibaba/qwen3.8-2.4t-a95b": alibaba.Qwen3_8_2_4t_A95bParams,
     "alibaba/qwen3.8-max": alibaba.Qwen3_8_MaxParams,
     "alibaba/qwq-plus": alibaba.Qwq_PlusParams,
     "anthropic/claude-3-5-haiku-20241022": anthropic.Claude_3_5_Haiku_20241022Params,

{modelparams-0.0.13 → modelparams-0.0.15}/src/modelparams/types/alibaba.py

@@ -23,6 +23,19 @@ Qwen_FlashParams = TypedDict(
 )
 setattr(Qwen_FlashParams, "__pydantic_config__", _PARAMS_CONFIG)
 
+Qwen_MaxParams = TypedDict(
+    "Qwen_MaxParams",
+    {
+        "max_tokens": Annotated[int, Field(ge=1)],
+        "temperature": Annotated[float, Field(ge=0, le=1.9)],
+        "top_p": Annotated[float, Field(ge=0, le=1)],
+        "extra_body.top_k": Annotated[int, Field(ge=1)],
+        "extra_body.chat_template_kwargs.enable_thinking": bool,
+    },
+    total=False,
+)
+setattr(Qwen_MaxParams, "__pydantic_config__", _PARAMS_CONFIG)
+
 Qwen_PlusParams = TypedDict(
     "Qwen_PlusParams",
     {
@@ -36,6 +49,19 @@ Qwen_PlusParams = TypedDict(
 )
 setattr(Qwen_PlusParams, "__pydantic_config__", _PARAMS_CONFIG)
 
+Qwen_TurboParams = TypedDict(
+    "Qwen_TurboParams",
+    {
+        "max_tokens": Annotated[int, Field(ge=1)],
+        "temperature": Annotated[float, Field(ge=0, le=1.9)],
+        "top_p": Annotated[float, Field(ge=0, le=1)],
+        "extra_body.top_k": Annotated[int, Field(ge=1)],
+        "extra_body.chat_template_kwargs.enable_thinking": bool,
+    },
+    total=False,
+)
+setattr(Qwen_TurboParams, "__pydantic_config__", _PARAMS_CONFIG)
+
 Qwen3_Coder_FlashParams = TypedDict(
     "Qwen3_Coder_FlashParams",
     {
@@ -141,6 +167,20 @@ Qwen3_7_PlusParams = TypedDict(
 )
 setattr(Qwen3_7_PlusParams, "__pydantic_config__", _PARAMS_CONFIG)
 
+Qwen3_8_2_4t_A95bParams = TypedDict(
+    "Qwen3_8_2_4t_A95bParams",
+    {
+        "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)],
+        "extra_body.top_k": Annotated[int, Field(ge=0)],
+        "extra_body.enable_thinking": bool,
+        "extra_body.thinking_budget": Annotated[int, Field(ge=1)],
+    },
+    total=False,
+)
+setattr(Qwen3_8_2_4t_A95bParams, "__pydantic_config__", _PARAMS_CONFIG)
+
 Qwen3_8_MaxParams = TypedDict(
     "Qwen3_8_MaxParams",
     {
@@ -169,7 +209,9 @@ setattr(Qwq_PlusParams, "__pydantic_config__", _PARAMS_CONFIG)
 
 __all__ = [
     "Qwen_FlashParams",
+    "Qwen_MaxParams",
     "Qwen_PlusParams",
+    "Qwen_TurboParams",
     "Qwen3_Coder_FlashParams",
     "Qwen3_Coder_PlusParams",
     "Qwen3_MaxParams",
@@ -178,6 +220,7 @@ __all__ = [
     "Qwen3_6_FlashParams",
     "Qwen3_7_MaxParams",
     "Qwen3_7_PlusParams",
+    "Qwen3_8_2_4t_A95bParams",
     "Qwen3_8_MaxParams",
     "Qwq_PlusParams",
 ]