Instructions to use MarsupialAI/Monstral-123B-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MarsupialAI/Monstral-123B-v2 with Transformers:

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

pipe = pipeline("text-generation", model="MarsupialAI/Monstral-123B-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MarsupialAI/Monstral-123B-v2")
model = AutoModelForCausalLM.from_pretrained("MarsupialAI/Monstral-123B-v2", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MarsupialAI/Monstral-123B-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MarsupialAI/Monstral-123B-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MarsupialAI/Monstral-123B-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MarsupialAI/Monstral-123B-v2

SGLang

How to use MarsupialAI/Monstral-123B-v2 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 "MarsupialAI/Monstral-123B-v2" \
    --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": "MarsupialAI/Monstral-123B-v2",
		"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 "MarsupialAI/Monstral-123B-v2" \
        --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": "MarsupialAI/Monstral-123B-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MarsupialAI/Monstral-123B-v2 with Docker Model Runner:
```
docker model run hf.co/MarsupialAI/Monstral-123B-v2
```

Metharme Chat Template?

by FenderBenderXXX - opened Jan 19, 2025

Discussion

FenderBenderXXX

Jan 19, 2025

•

edited Jan 19, 2025

I'm sorry, I swear I've been looking literally for cumulative hours before coming here to ask such a basic question, but what is--or at least where can I find a chat template for Metharme? I found examples of how TO prompt on the HuggingFace page, but I'm still super new to this and don't understand how to take that knowledge and make a template for my GUI--which is LM Studio currently but I also have Jan.ai, Backyard AI, and SillyTavern set up with a llama.cpp backend. ST accepts the preset for Methception v1.4, but I really don't like working directly in the llama.cpp backend. I barely figured out how to change context, GPU layer, and threads.

FenderBenderXXX changed discussion title from Metharme Prompt Template? to Metharme Chat Template? Jan 19, 2025

lightning-missile

Jan 21, 2025

•

edited Jan 21, 2025

<|system|>{system prompt goes here}<|user|>{user prompt goes here}<|model|>{model's response goes here}

Since you have silly tavern, you can import settings like this https://huggingface.co/Konnect1221/Methception-Llamaception-SillyTavern-Preset

FenderBenderXXX

Jan 22, 2025

<|system|>{system prompt goes here}<|user|>{user prompt goes here}<|model|>{model's response goes here}

Since you have silly tavern, you can import settings like this https://huggingface.co/Konnect1221/Methception-Llamaception-SillyTavern-Preset

Thank you thank you thank you~~

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