feat(batch): support for mistral api

[vcharvet]

Supports for calling MistralAI batch API. The modification does not impact other aspects of the library.

Usage is similar as openai/anthropic

from pydantic import BaseModel

class User(BaseModel):
    name: str
    age: int

processor = BatchProcessor("mistral/mistral-small-latest", User)

Currently only support mistralai<2.0.0 as mentionned in #2137
Tested with instructor==1.14.5

[vcharvet]

Mistral provider has been tested on the same snippet as https://python.useinstructor.com/concepts/batch/#in-memory-processing:

import time

from instructor.batch import BatchProcessor, extract_results
from pydantic import BaseModel


class User(BaseModel):
    name: str
    age: int


processor = BatchProcessor("mistral/mistral-small-latest", User)


messages_list = [
    [
        {"role": "system", "content": "Extract user information from text."},
        {"role": "user", "content": "Hi, I'm Alice and I'm 28 years old."},
    ],
    [
        {"role": "system", "content": "Extract user information from text."},
        {"role": "user", "content": "Hello, I'm Bob, 35 years old."},
    ],
]

# Create batch file
processor.create_batch_from_messages(
    file_path="batch_requests.jsonl",
    messages_list=messages_list,
    max_tokens=100000,
    temperature=0.1,
)

# Submit batch job
batch_id = processor.submit_batch("batch_requests.jsonl")
print(f"Batch job submitted: {batch_id}")

# Check status and retrieve results
status = processor.get_batch_status(batch_id)
while status["status"] not in ["SUCCESS", "FAILED", "CANCELLED"]:
    status = processor.get_batch_status(batch_id)
    time.sleep(10)

print(f"Job status {status['status']}")

all_results = processor.retrieve_results(batch_id)
extracted_results = extract_results(all_results)
for user in extracted_results:
    print(f"Name: {user.name}, Age: {user.age}")
# Name: Alice, Age: 28
# Name: Bob,  Age: 35