Instructions to use Qiskit/mistral-small-3.2-24b-qiskit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qiskit/mistral-small-3.2-24b-qiskit with Transformers:

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

pipe = pipeline("text-generation", model="Qiskit/mistral-small-3.2-24b-qiskit")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Qiskit/mistral-small-3.2-24b-qiskit")
model = AutoModelForMultimodalLM.from_pretrained("Qiskit/mistral-small-3.2-24b-qiskit", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Qiskit/mistral-small-3.2-24b-qiskit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Install mistral-common:
pip install --upgrade mistral-common
# Start the vLLM server:
vllm serve "Qiskit/mistral-small-3.2-24b-qiskit" --tokenizer_mode mistral --config_format mistral --load_format mistral --tool-call-parser mistral --enable-auto-tool-choice
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Qiskit/mistral-small-3.2-24b-qiskit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Qiskit/mistral-small-3.2-24b-qiskit

SGLang

How to use Qiskit/mistral-small-3.2-24b-qiskit 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 "Qiskit/mistral-small-3.2-24b-qiskit" \
    --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": "Qiskit/mistral-small-3.2-24b-qiskit",
		"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 "Qiskit/mistral-small-3.2-24b-qiskit" \
        --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": "Qiskit/mistral-small-3.2-24b-qiskit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Qiskit/mistral-small-3.2-24b-qiskit with Docker Model Runner:
```
docker model run hf.co/Qiskit/mistral-small-3.2-24b-qiskit
```

mistral-small-3.2-24b-qiskit / params.json

cbjuan

Add model card + params file (#2)

46ebec6 verified 11 months ago

Raw

History Blame

807 Bytes

	{
	"dim": 5120,
	"n_layers": 40,
	"head_dim": 128,
	"hidden_dim": 32768,
	"n_heads": 32,
	"n_kv_heads": 8,
	"rope_theta": 1000000000.0,
	"norm_eps": 1e-05,
	"vocab_size": 131072,
	"max_position_embeddings": 131072,
	"vision_encoder": {
	"hidden_size": 1024,
	"num_channels": 3,
	"max_image_size": 1540,
	"patch_size": 14,
	"rope_theta": 10000.0,
	"intermediate_size": 4096,
	"num_hidden_layers": 24,
	"num_attention_heads": 16,
	"adapter_bias": false,
	"mm_projector_id": "patch_merge",
	"spatial_merge_size": 2,
	"add_pre_mm_projector_layer_norm": true,
	"image_token_id": 10,
	"image_break_token_id": 12,
	"image_end_token_id": 13,
	"image_size": 1540
	}
	}