Let's see if the Ministral 3 model knows when to pick a fight !

from typing import Any

from openai import OpenAI
from huggingfacehub import hfhub_download

Modify OpenAI's API key and API base to use vLLM's API server.
openaiapikey = "EMPTY"
openaiapibase = "http://localhost:8000/v1"

TEMP = 0.7
TOP_P = 0.95
MAX_TOK = 262144
client = OpenAI(
    apikey=openaiapi_key,
    baseurl=openaiapi_base,
)

models = client.models.list()
model = models.data[0].id

def loadsystemprompt(repo_id: str, filename: str) -> dict[str, Any]:
    filepath = hfhubdownload(repoid=repo_id, filename=filename)
    with open(file_path, "r") as file:
        system_prompt = file.read()

indexbeginthink = system_prompt.find("[THINK]")
    indexendthink = system_prompt.find("[/THINK]")

return {
        "role": "system",
        "content": [
            {"type": "text", "text": systemprompt[:indexbegin_think]},
            {
                "type": "thinking",
                "thinking": system_prompt[
                    indexbeginthink + len("[THINK]") : indexendthink
                ],
                "closed": True,
            },
            {
                "type": "text",
                "text": systemprompt[indexend_think + len("[/THINK]") :],
            },
        ],
    }

SYSTEMPROMPT = loadsystemprompt(model, "SYSTEMPROMPT.txt")

image_url = "https://static.wikia.nocookie.net/essentialsdocs/images/7/70/Battle.png/revision/latest?cb=20220523172438"

messages = [
    SYSTEM_PROMPT,
    {
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": "What action do you think I should take in this situation? List all the possible actions and explain why you think they are good or bad.",
            },
            {"type": "imageurl", "imageurl": {"url": image_url}},
        ],
    },
]

stream = client.chat.completions.create(
    model=model,
    messages=messages,
    stream=True,
    temperature=TEMP,
    topp=TOPP,
    maxtokens=MAXTOK,
)

print("client: Start streaming chat completions...:\n")
printedreasoningcontent = False
answer = []

for chunk in stream:
    reasoning_content = None
    content = None
    # Check the content is reasoning_content or content
    if hasattr(chunk.choices[0].delta, "reasoning_content"):
        reasoningcontent = chunk.choices[0].delta.reasoningcontent
    if hasattr(chunk.choices[0].delta, "content"):
        content = chunk.choices[0].delta.content

if reasoning_content is not None:
        if not printedreasoningcontent:
            printedreasoningcontent = True
            print("Start reasoning:\n", end="", flush=True)
        print(reasoning_content, end="", flush=True)
    elif content is not None:
        # Extract and print the content
        if not reasoningcontent and printedreasoning_content:
            answer.extend(content)
        print(content, end="", flush=True)

if answer:
    print("\n\n=============\nAnswer\n=============\n")
    print("".join(answer))
else:
    print("\n\n=============\nNo Answer\n=============\n")
    print(
        "No answer was generated by the model, probably because the maximum number of tokens was reached."
    )

Now we'll make it compute some maths !

from typing import Any

from openai import OpenAI
from huggingfacehub import hfhub_download

Modify OpenAI's API key and API base to use vLLM's API server.
openaiapikey = "EMPTY"
openaiapibase = "http://localhost:8000/v1"

TEMP = 0.7
TOP_P = 0.95
MAX_TOK = 262144
client = OpenAI(
    apikey=openaiapi_key,
    baseurl=openaiapi_base,
)

models = client.models.list()
model = models.data[0].id

def loadsystemprompt(repo_id: str, filename: str) -> dict[str, Any]:
    filepath = hfhubdownload(repoid=repo_id, filename=filename)
    with open(file_path, "r") as file:
        system_prompt = file.read()

indexbeginthink = system_prompt.find("[THINK]")
    indexendthink = system_prompt.find("[/THINK]")

return {
        "role": "system",
        "content": [
            {"type": "text", "text": systemprompt[:indexbegin_think]},
            {
                "type": "thinking",
                "thinking": system_prompt[
                    indexbeginthink + len("[THINK]") : indexendthink
                ],
                "closed": True,
            },
            {
                "type": "text",
                "text": systemprompt[indexend_think + len("[/THINK]") :],
            },
        ],
    }

SYSTEMPROMPT = loadsystemprompt(model, "SYSTEMPROMPT.txt")

image_url = "https://i.ytimg.com/vi/5Y3xLHeyKZU/hqdefault.jpg"

messages = [
    SYSTEM_PROMPT,
    {
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": "Solve the equations. If they contain only numbers, use your calculator, else only think. Answer in the language of the image.",
            },
            {"type": "imageurl", "imageurl": {"url": image_url}},
        ],
    },
]

stream = client.chat.completions.create(
    model=model,
    messages=messages,
    stream=True,
    temperature=TEMP,
    topp=TOPP,
    maxtokens=MAXTOK,
)

print("client: Start streaming chat completions...:\n")
printedreasoningcontent = False
answer = []

for chunk in stream:
    reasoning_content = None
    content = None
    # Check the content is reasoning_content or content
    if hasattr(chunk.choices[0].delta, "reasoning_content"):
        reasoningcontent = chunk.choices[0].delta.reasoningcontent
    if hasattr(chunk.choices[0].delta, "content"):
        content = chunk.choices[0].delta.content

if reasoning_content is not None:
        if not printedreasoningcontent:
            printedreasoningcontent = True
            print("Start reasoning:\n", end="", flush=True)
        print(reasoning_content, end="", flush=True)
    if content is not None:
        # Extract and print the content
        if not reasoningcontent and printedreasoning_content:
            answer.extend(content)
        print(content, end="", flush=True)

if answer:
    print("\n\n=============\nAnswer\n=============\n")
    print("".join(answer))
else:
    print("\n\n=============\nNo Answer\n=============\n")
    print(
        "No answer was generated by the model, probably because the maximum number of tokens was reached."
    )

Let's do more maths and leave it up to the model to figure out how to achieve a result.

from typing import Any
from openai import OpenAI
from huggingfacehub import hfhub_download

Modify OpenAI's API key and API base to use vLLM's API server.
openaiapikey = "EMPTY"
openaiapibase = "http://localhost:8000/v1"

TEMP = 0.7
TOP_P = 0.95
MAX_TOK = 262144
client = OpenAI(
    apikey=openaiapi_key,
    baseurl=openaiapi_base,
)

models = client.models.list()
model = models.data[0].id

def loadsystemprompt(repo_id: str, filename: str) -> dict[str, Any]:
    filepath = hfhubdownload(repoid=repo_id, filename=filename)
    with open(file_path, "r") as file:
        system_prompt = file.read()

indexbeginthink = system_prompt.find("[THINK]")
    indexendthink = system_prompt.find("[/THINK]")

return {
        "role": "system",
        "content": [
            {"type": "text", "text": systemprompt[:indexbegin_think]},
            {
                "type": "thinking",
                "thinking": system_prompt[
                    indexbeginthink + len("[THINK]") : indexendthink
                ],
                "closed": True,
            },
            {
                "type": "text",
                "text": systemprompt[indexend_think + len("[/THINK]") :],
            },
        ],
    }

SYSTEMPROMPT = loadsystemprompt(model, "SYSTEMPROMPT.txt")

query = "Use each number in 2,5,6,3 exactly once, along with any combination of +, -, ×, ÷ (and parentheses for grouping), to make the number 24."

messages = [
    SYSTEM_PROMPT,
    {"role": "user", "content": query}
]
stream = client.chat.completions.create(
  model=model,
  messages=messages,
  stream=True,
  temperature=TEMP,
  topp=TOPP,
  maxtokens=MAXTOK,
)

print("client: Start streaming chat completions...:\n")
printedreasoningcontent = False
answer = []

for chunk in stream:
    reasoning_content = None
    content = None
    # Check the content is reasoning_content or content
    if hasattr(chunk.choices[0].delta, "reasoning_content"):
        reasoningcontent = chunk.choices[0].delta.reasoningcontent
    if hasattr(chunk.choices[0].delta, "content"):
        content = chunk.choices[0].delta.content

if reasoning_content is not None:
        if not printedreasoningcontent:
            printedreasoningcontent = True
            print("Start reasoning:\n", end="", flush=True)
        print(reasoning_content, end="", flush=True)
    if content is not None:
        # Extract and print the content
        if not reasoningcontent and printedreasoning_content:
            answer.extend(content)
        print(content, end="", flush=True)

if answer:
    print("\n\n=============\nAnswer\n=============\n")
    print("".join(answer))
else:
    print("\n\n=============\nNo Answer\n=============\n")
    print("No answer was generated by the model, probably because the maximum number of tokens was reached.")

import torch
from transformers import Mistral3ForConditionalGeneration, MistralCommonBackend

model_id = "mistralai/Ministral-3-14B-Reasoning-2512"

tokenizer = MistralCommonBackend.frompretrained(modelid)
model = Mistral3ForConditionalGeneration.from_pretrained(
    modelid, torchdtype=torch.bfloat16, device_map="auto"
)

image_url = "https://static.wikia.nocookie.net/essentialsdocs/images/7/70/Battle.png/revision/latest?cb=20220523172438"

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "text",
                "text": "What action do you think I should take in this situation? List all the possible actions and explain why you think they are good or bad.",
            },
            {"type": "imageurl", "imageurl": {"url": image_url}},
        ],
    },
]

tokenized = tokenizer.applychattemplate(messages, returntensors="pt", returndict=True)

tokenized["inputids"] = tokenized["inputids"].to(device="cuda")
tokenized["pixelvalues"] = tokenized["pixelvalues"].to(dtype=torch.bfloat16, device="cuda")
imagesizes = [tokenized["pixelvalues"].shape[-2:]]

output = model.generate(
    tokenized,
    imagesizes=imagesizes,
    maxnewtokens=8092,
)[0]

decodedoutput = tokenizer.decode(output[len(tokenized["inputids"][0]):])
print(decoded_output)