Instructions to use stelterlab/Mistral-Small-24B-Instruct-2501-AWQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use stelterlab/Mistral-Small-24B-Instruct-2501-AWQ with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="stelterlab/Mistral-Small-24B-Instruct-2501-AWQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("stelterlab/Mistral-Small-24B-Instruct-2501-AWQ")
model = AutoModelForCausalLM.from_pretrained("stelterlab/Mistral-Small-24B-Instruct-2501-AWQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use stelterlab/Mistral-Small-24B-Instruct-2501-AWQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "stelterlab/Mistral-Small-24B-Instruct-2501-AWQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "stelterlab/Mistral-Small-24B-Instruct-2501-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/stelterlab/Mistral-Small-24B-Instruct-2501-AWQ

SGLang

How to use stelterlab/Mistral-Small-24B-Instruct-2501-AWQ with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "stelterlab/Mistral-Small-24B-Instruct-2501-AWQ" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "stelterlab/Mistral-Small-24B-Instruct-2501-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "stelterlab/Mistral-Small-24B-Instruct-2501-AWQ" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "stelterlab/Mistral-Small-24B-Instruct-2501-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use stelterlab/Mistral-Small-24B-Instruct-2501-AWQ with Docker Model Runner:
```
docker model run hf.co/stelterlab/Mistral-Small-24B-Instruct-2501-AWQ
```

Tool Calling issue with stelterlab/Mistral-Small-24B-Instruct-2501-AWQ

by sbhatt765 - opened May 14, 2025

Discussion

sbhatt765

May 14, 2025

•

edited May 14, 2025

I am getting 400 while trying to get tool call working with this one -

Here is the docker config-
sudo docker run --runtime nvidia --gpus all
-v /data/Mistral-Small-24B-Instruct-2501-AWQ:/models/Mistral-Small-24B-Instruct-2501-AWQ
-p 8085:8085
--ipc=host
vllm/vllm-openai:latest
--model /models/Mistral-Small-24B-Instruct-2501-AWQ
--gpu-memory-utilization 0.85
--max-model-len 4096
--tokenizer-mode mistral
--tool-call-parser mistral
--enable-auto-tool-choice
--enforce-eager
--dtype half

let me know if the vLLM config looks alright, I am not using the chat-template flag(getting same error with that as well).

model_client = OpenAIChatCompletionClient(

 model="/models/Mistral-Small-24B-Instruct-2501-AWQ",
 base_url="http://localhost:8085/v1",

api_key="EMPTY",
model_info={
        "vision": False,
        "function_calling": True,
        "json_output": True,
        "family": "unknown",
    },

)

stelterlab

Owner May 18, 2025

•

edited May 18, 2025

Hi!

You might get faster response when you are asking at the source (of the models) or on the discussions of vLLM.

Try to call it with:

--tokenizer mistralai/Mistral-Small-24B-Instruct-2501

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment