mistralai 0.1.8__tar.gz → 0.3.0__tar.gz

{mistralai-0.1.8 → mistralai-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mistralai
-Version: 0.1.8
+Version: 0.3.0
 Summary: 
 Author: Bam4d
 Author-email: bam4d@mistral.ai
@@ -10,7 +10,7 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
-Requires-Dist: httpx (>=0.25.2,<0.26.0)
+Requires-Dist: httpx (>=0.25.2,<1)
 Requires-Dist: orjson (>=3.9.10,<4.0.0)
 Requires-Dist: pydantic (>=2.5.2,<3.0.0)
 Description-Content-Type: text/markdown

{mistralai-0.1.8 → mistralai-0.3.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "mistralai"
-version = "0.1.8"
+version = "0.3.0"
 description = ""
 authors = ["Bam4d <bam4d@mistral.ai>"]
 readme = "README.md"
@@ -26,7 +26,7 @@ exclude = ["docs", "tests", "examples", "tools", "build"]
 python = "^3.9"
 orjson = "^3.9.10"
 pydantic = "^2.5.2"
-httpx = "^0.25.2"
+httpx = ">= 0.25.2, < 1"
 
 
 [tool.poetry.group.dev.dependencies]

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/async_client.py

@@ -1,6 +1,5 @@
-import os
+import asyncio
 import posixpath
-import time
 from json import JSONDecodeError
 from typing import Any, AsyncGenerator, Dict, List, Optional, Union
 
@@ -34,7 +33,7 @@ from mistralai.models.models import ModelList
 class MistralAsyncClient(ClientBase):
     def __init__(
         self,
-        api_key: Optional[str] = os.environ.get("MISTRAL_API_KEY", None),
+        api_key: Optional[str] = None,
         endpoint: str = ENDPOINT,
         max_retries: int = 5,
         timeout: int = 120,
@@ -93,7 +92,7 @@ class MistralAsyncClient(ClientBase):
     async def _request(
         self,
         method: str,
-        json: Dict[str, Any],
+        json: Optional[Dict[str, Any]],
         path: str,
         stream: bool = False,
         attempt: int = 1,
@@ -151,7 +150,7 @@ class MistralAsyncClient(ClientBase):
             if attempt > self._max_retries:
                 raise MistralAPIStatusException.from_response(response, message=str(e)) from e
             backoff = 2.0**attempt  # exponential backoff
-            time.sleep(backoff)
+            await asyncio.sleep(backoff)
 
             # Retry as a generator
             async for r in self._request(method, json, path, stream=stream, attempt=attempt):
@@ -292,3 +291,74 @@ class MistralAsyncClient(ClientBase):
             return ModelList(**response)
 
         raise MistralException("No response received")
+
+    async def completion(
+        self,
+        model: str,
+        prompt: str,
+        suffix: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        top_p: Optional[float] = None,
+        random_seed: Optional[int] = None,
+        stop: Optional[List[str]] = None,
+    ) -> ChatCompletionResponse:
+        """An asynchronous completion endpoint that returns a single response.
+
+        Args:
+            model (str): model the name of the model to get completions with, e.g. codestral-latest
+            prompt (str): the prompt to complete
+            suffix (Optional[str]): the suffix to append to the prompt for fill-in-the-middle completion
+            temperature (Optional[float], optional): temperature the temperature to use for sampling, e.g. 0.5.
+            max_tokens (Optional[int], optional): the maximum number of tokens to generate, e.g. 100. Defaults to None.
+            top_p (Optional[float], optional): the cumulative probability of tokens to generate, e.g. 0.9.
+            Defaults to None.
+            random_seed (Optional[int], optional): the random seed to use for sampling, e.g. 42. Defaults to None.
+            stop (Optional[List[str]], optional): a list of tokens to stop generation at, e.g. ['/n/n']
+        Returns:
+            Dict[str, Any]: a response object containing the generated text.
+        """
+        request = self._make_completion_request(
+            prompt, model, suffix, temperature, max_tokens, top_p, random_seed, stop
+        )
+        single_response = self._request("post", request, "v1/fim/completions")
+
+        async for response in single_response:
+            return ChatCompletionResponse(**response)
+
+        raise MistralException("No response received")
+
+    async def completion_stream(
+        self,
+        model: str,
+        prompt: str,
+        suffix: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        top_p: Optional[float] = None,
+        random_seed: Optional[int] = None,
+        stop: Optional[List[str]] = None,
+    ) -> AsyncGenerator[ChatCompletionStreamResponse, None]:
+        """An asynchronous completion endpoint that returns a streaming response.
+
+        Args:
+            model (str): model the name of the model to get completions with, e.g. codestral-latest
+            prompt (str): the prompt to complete
+            suffix (Optional[str]): the suffix to append to the prompt for fill-in-the-middle completion
+            temperature (Optional[float], optional): temperature the temperature to use for sampling, e.g. 0.5.
+            max_tokens (Optional[int], optional): the maximum number of tokens to generate, e.g. 100. Defaults to None.
+            top_p (Optional[float], optional): the cumulative probability of tokens to generate, e.g. 0.9.
+            Defaults to None.
+            random_seed (Optional[int], optional): the random seed to use for sampling, e.g. 42. Defaults to None.
+            stop (Optional[List[str]], optional): a list of tokens to stop generation at, e.g. ['/n/n']
+
+        Returns:
+            Dict[str, Any]: a response object containing the generated text.
+        """
+        request = self._make_completion_request(
+            prompt, model, suffix, temperature, max_tokens, top_p, random_seed, stop, stream=True
+        )
+        async_response = self._request("post", request, "v1/fim/completions", stream=True)
+
+        async for json_response in async_response:
+            yield ChatCompletionStreamResponse(**json_response)

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/client.py

@@ -1,4 +1,3 @@
-import os
 import posixpath
 import time
 from json import JSONDecodeError
@@ -31,7 +30,7 @@ class MistralClient(ClientBase):
 
     def __init__(
         self,
-        api_key: Optional[str] = os.environ.get("MISTRAL_API_KEY", None),
+        api_key: Optional[str] = None,
         endpoint: str = ENDPOINT,
         max_retries: int = 5,
         timeout: int = 120,
@@ -86,7 +85,7 @@ class MistralClient(ClientBase):
     def _request(
         self,
         method: str,
-        json: Dict[str, Any],
+        json: Optional[Dict[str, Any]],
         path: str,
         stream: bool = False,
         attempt: int = 1,
@@ -286,3 +285,77 @@ class MistralClient(ClientBase):
             return ModelList(**response)
 
         raise MistralException("No response received")
+
+    def completion(
+        self,
+        model: str,
+        prompt: str,
+        suffix: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        top_p: Optional[float] = None,
+        random_seed: Optional[int] = None,
+        stop: Optional[List[str]] = None,
+    ) -> ChatCompletionResponse:
+        """A completion endpoint that returns a single response.
+
+        Args:
+            model (str): model the name of the model to get completion with, e.g. codestral-latest
+            prompt (str): the prompt to complete
+            suffix (Optional[str]): the suffix to append to the prompt for fill-in-the-middle completion
+            temperature (Optional[float], optional): temperature the temperature to use for sampling, e.g. 0.5.
+            max_tokens (Optional[int], optional): the maximum number of tokens to generate, e.g. 100. Defaults to None.
+            top_p (Optional[float], optional): the cumulative probability of tokens to generate, e.g. 0.9.
+            Defaults to None.
+            random_seed (Optional[int], optional): the random seed to use for sampling, e.g. 42. Defaults to None.
+            stop (Optional[List[str]], optional): a list of tokens to stop generation at, e.g. ['/n/n']
+
+        Returns:
+            Dict[str, Any]: a response object containing the generated text.
+        """
+        request = self._make_completion_request(
+            prompt, model, suffix, temperature, max_tokens, top_p, random_seed, stop
+        )
+
+        single_response = self._request("post", request, "v1/fim/completions", stream=False)
+
+        for response in single_response:
+            return ChatCompletionResponse(**response)
+
+        raise MistralException("No response received")
+
+    def completion_stream(
+        self,
+        model: str,
+        prompt: str,
+        suffix: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        top_p: Optional[float] = None,
+        random_seed: Optional[int] = None,
+        stop: Optional[List[str]] = None,
+    ) -> Iterable[ChatCompletionStreamResponse]:
+        """An asynchronous completion endpoint that streams responses.
+
+        Args:
+            model (str): model the name of the model to get completions with, e.g. codestral-latest
+            prompt (str): the prompt to complete
+            suffix (Optional[str]): the suffix to append to the prompt for fill-in-the-middle completion
+            temperature (Optional[float], optional): temperature the temperature to use for sampling, e.g. 0.5.
+            max_tokens (Optional[int], optional): the maximum number of tokens to generate, e.g. 100. Defaults to None.
+            top_p (Optional[float], optional): the cumulative probability of tokens to generate, e.g. 0.9.
+            Defaults to None.
+            random_seed (Optional[int], optional): the random seed to use for sampling, e.g. 42. Defaults to None.
+            stop (Optional[List[str]], optional): a list of tokens to stop generation at, e.g. ['/n/n']
+
+        Returns:
+            Iterable[Dict[str, Any]]: a generator that yields response objects containing the generated text.
+        """
+        request = self._make_completion_request(
+            prompt, model, suffix, temperature, max_tokens, top_p, random_seed, stop, stream=True
+        )
+
+        response = self._request("post", request, "v1/fim/completions", stream=True)
+
+        for json_streamed_response in response:
+            yield ChatCompletionStreamResponse(**json_streamed_response)

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/client_base.py

@@ -10,10 +10,7 @@ from mistralai.exceptions import (
 )
 from mistralai.models.chat_completion import ChatMessage, Function, ResponseFormat, ToolChoice
 
-logging.basicConfig(
-    format="%(asctime)s %(levelname)s %(name)s: %(message)s",
-    level=os.getenv("LOG_LEVEL", "ERROR"),
-)
+CLIENT_VERSION = "0.2.0"
 
 
 class ClientBase(ABC):
@@ -27,16 +24,19 @@ class ClientBase(ABC):
         self._max_retries = max_retries
         self._timeout = timeout
 
-        self._endpoint = endpoint
+        if api_key is None:
+            api_key = os.environ.get("MISTRAL_API_KEY")
+        if api_key is None:
+            raise MistralException(message="API key not provided. Please set MISTRAL_API_KEY environment variable.")
         self._api_key = api_key
+        self._endpoint = endpoint
         self._logger = logging.getLogger(__name__)
 
         # For azure endpoints, we default to the mistral model
         if "inference.azure.com" in self._endpoint:
             self._default_model = "mistral"
 
-        # This should be automatically updated by the deploy script
-        self._version = "0.1.8"
+        self._version = CLIENT_VERSION
 
     def _parse_tools(self, tools: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
         parsed_tools: List[Dict[str, Any]] = []
@@ -73,6 +73,63 @@ class ClientBase(ABC):
 
         return parsed_messages
 
+    def _make_completion_request(
+        self,
+        prompt: str,
+        model: Optional[str] = None,
+        suffix: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        top_p: Optional[float] = None,
+        random_seed: Optional[int] = None,
+        stop: Optional[List[str]] = None,
+        stream: Optional[bool] = False,
+    ) -> Dict[str, Any]:
+        request_data: Dict[str, Any] = {
+            "prompt": prompt,
+            "suffix": suffix,
+            "model": model,
+            "stream": stream,
+        }
+
+        if stop is not None:
+            request_data["stop"] = stop
+
+        if model is not None:
+            request_data["model"] = model
+        else:
+            if self._default_model is None:
+                raise MistralException(message="model must be provided")
+            request_data["model"] = self._default_model
+
+        request_data.update(
+            self._build_sampling_params(
+                temperature=temperature, max_tokens=max_tokens, top_p=top_p, random_seed=random_seed
+            )
+        )
+
+        self._logger.debug(f"Completion request: {request_data}")
+
+        return request_data
+
+    def _build_sampling_params(
+        self,
+        max_tokens: Optional[int],
+        random_seed: Optional[int],
+        temperature: Optional[float],
+        top_p: Optional[float],
+    ) -> Dict[str, Any]:
+        params = {}
+        if temperature is not None:
+            params["temperature"] = temperature
+        if max_tokens is not None:
+            params["max_tokens"] = max_tokens
+        if top_p is not None:
+            params["top_p"] = top_p
+        if random_seed is not None:
+            params["random_seed"] = random_seed
+        return params
+
     def _make_chat_request(
         self,
         messages: List[Any],
@@ -99,16 +156,14 @@ class ClientBase(ABC):
                 raise MistralException(message="model must be provided")
             request_data["model"] = self._default_model
 
+        request_data.update(
+            self._build_sampling_params(
+                temperature=temperature, max_tokens=max_tokens, top_p=top_p, random_seed=random_seed
+            )
+        )
+
         if tools is not None:
             request_data["tools"] = self._parse_tools(tools)
-        if temperature is not None:
-            request_data["temperature"] = temperature
-        if max_tokens is not None:
-            request_data["max_tokens"] = max_tokens
-        if top_p is not None:
-            request_data["top_p"] = top_p
-        if random_seed is not None:
-            request_data["random_seed"] = random_seed
         if stream is not None:
             request_data["stream"] = stream
 

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/constants.py

@@ -1,5 +1,3 @@
-
-
 RETRY_STATUS_CODES = {429, 500, 502, 503, 504}
 
 ENDPOINT = "https://api.mistral.ai"

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/exceptions.py

@@ -35,9 +35,7 @@ class MistralAPIException(MistralException):
         self.headers = headers or {}
 
     @classmethod
-    def from_response(
-        cls, response: Response, message: Optional[str] = None
-    ) -> MistralAPIException:
+    def from_response(cls, response: Response, message: Optional[str] = None) -> MistralAPIException:
         return cls(
             message=message or response.text,
             http_status=response.status_code,
@@ -47,8 +45,10 @@ class MistralAPIException(MistralException):
     def __repr__(self) -> str:
         return f"{self.__class__.__name__}(message={str(self)}, http_status={self.http_status})"
 
+
 class MistralAPIStatusException(MistralAPIException):
     """Returned when we receive a non-200 response from the API that we should retry"""
 
+
 class MistralConnectionException(MistralException):
     """Returned when the SDK can not reach the API server for any reason"""

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/models/chat_completion.py

@@ -47,6 +47,7 @@ class ChatMessage(BaseModel):
     content: Union[str, List[str]]
     name: Optional[str] = None
     tool_calls: Optional[List[ToolCall]] = None
+    tool_call_id: Optional[str] = None
 
 
 class DeltaMessage(BaseModel):

{mistralai-0.1.8 → mistralai-0.3.0}/src/mistralai/models/models.py

@@ -17,6 +17,7 @@ class ModelPermission(BaseModel):
     group: Optional[str] = None
     is_blocking: bool = False
 
+
 class ModelCard(BaseModel):
     id: str
     object: str