feat(model-profiles): distribute data across packages (#34024)

libs/partners/mistralai/langchain_mistralai/chat_models.py

@@ -23,7 +23,11 @@ from langchain_core.callbacks import (
     AsyncCallbackManagerForLLMRun,
     CallbackManagerForLLMRun,
 )
-from langchain_core.language_models import LanguageModelInput
+from langchain_core.language_models import (
+    LanguageModelInput,
+    ModelProfile,
+    ModelProfileRegistry,
+)
 from langchain_core.language_models.chat_models import BaseChatModel, LangSmithParams
 from langchain_core.language_models.llms import create_base_retry_decorator
 from langchain_core.messages import (
@@ -70,6 +74,7 @@ from pydantic import (
 from typing_extensions import Self
 
 from langchain_mistralai._compat import _convert_from_v1_to_mistral
+from langchain_mistralai.data._profiles import _PROFILES
 
 if TYPE_CHECKING:
     from collections.abc import AsyncIterator, Iterator, Sequence
@@ -86,6 +91,14 @@ TOOL_CALL_ID_PATTERN = re.compile(r"^[a-zA-Z0-9]{9}$")
 global_ssl_context = ssl.create_default_context(cafile=certifi.where())
 
 
+_MODEL_PROFILES = cast("ModelProfileRegistry", _PROFILES)
+
+
+def _get_default_model_profile(model_name: str) -> ModelProfile:
+    default = _MODEL_PROFILES.get(model_name) or {}
+    return default.copy()
+
+
 def _create_retry_decorator(
     llm: ChatMistralAI,
     run_manager: AsyncCallbackManagerForLLMRun | CallbackManagerForLLMRun | None = None,
@@ -632,6 +645,13 @@ class ChatMistralAI(BaseChatModel):
 
         return self
 
+    @model_validator(mode="after")
+    def _set_model_profile(self) -> Self:
+        """Set model profile if not overridden."""
+        if self.profile is None:
+            self.profile = _get_default_model_profile(self.model)
+        return self
+
     def _generate(
         self,
         messages: list[BaseMessage],

libs/partners/mistralai/langchain_mistralai/data/__init__.py

@@ -0,0 +1 @@
+"""Model profile data. All edits should be made in profile_augmentations.toml."""

libs/partners/mistralai/langchain_mistralai/data/_profiles.py

@@ -0,0 +1,247 @@
+"""Auto-generated model profiles.
+
+DO NOT EDIT THIS FILE MANUALLY.
+This file is generated by the langchain-profiles CLI tool.
+
+It contains data derived from the models.dev project.
+
+Source: https://github.com/sst/models.dev
+License: MIT License
+
+To update these data, refer to the instructions here:
+
+https://docs.langchain.com/oss/python/langchain/models#updating-or-overwriting-profile-data
+"""
+
+from typing import Any
+
+_PROFILES: dict[str, dict[str, Any]] = {
+    "devstral-medium-2507": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "open-mixtral-8x22b": {
+        "max_input_tokens": 64000,
+        "max_output_tokens": 64000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "ministral-8b-latest": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "pixtral-large-latest": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": True,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "ministral-3b-latest": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "pixtral-12b": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": True,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "mistral-medium-2505": {
+        "max_input_tokens": 131072,
+        "max_output_tokens": 131072,
+        "image_inputs": True,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "devstral-small-2505": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "mistral-medium-2508": {
+        "max_input_tokens": 262144,
+        "max_output_tokens": 262144,
+        "image_inputs": True,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "mistral-small-latest": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "image_inputs": True,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "magistral-small": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": True,
+        "tool_calling": True,
+    },
+    "devstral-small-2507": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "codestral-latest": {
+        "max_input_tokens": 256000,
+        "max_output_tokens": 4096,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "open-mixtral-8x7b": {
+        "max_input_tokens": 32000,
+        "max_output_tokens": 32000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "mistral-nemo": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 128000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "open-mistral-7b": {
+        "max_input_tokens": 8000,
+        "max_output_tokens": 8000,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "mistral-large-latest": {
+        "max_input_tokens": 131072,
+        "max_output_tokens": 16384,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "mistral-medium-latest": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "image_inputs": True,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": False,
+        "tool_calling": True,
+    },
+    "magistral-medium-latest": {
+        "max_input_tokens": 128000,
+        "max_output_tokens": 16384,
+        "image_inputs": False,
+        "audio_inputs": False,
+        "video_inputs": False,
+        "image_outputs": False,
+        "audio_outputs": False,
+        "video_outputs": False,
+        "reasoning_output": True,
+        "tool_calling": True,
+    },
+}