Instructions to use PleIAs/Monad with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PleIAs/Monad with Transformers:

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

pipe = pipeline("text-generation", model="PleIAs/Monad")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("PleIAs/Monad")
model = AutoModelForCausalLM.from_pretrained("PleIAs/Monad")
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 Settings

vLLM

How to use PleIAs/Monad with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/PleIAs/Monad

SGLang

How to use PleIAs/Monad 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 "PleIAs/Monad" \
    --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": "PleIAs/Monad",
		"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 "PleIAs/Monad" \
        --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": "PleIAs/Monad",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use PleIAs/Monad with Docker Model Runner:
```
docker model run hf.co/PleIAs/Monad
```

Evaluation Results for PleIAs/Monad & Suggested Model Card Update

by GODELEV - opened 3 days ago

Discussion

GODELEV

3 days ago

TO: PleIAs Team

DATE: June 2, 2026

SUBJECT: Evaluation Results for PleIAs/Monad & Suggested Model Card Update

First of all, I wanted to reach out and congratulate you on building such a fabulous model! The work you've done with PleIAs/Monad is truly impressive, and it's exciting to see its capabilities in action.

I have recently run a comprehensive evaluation suite on the model. I wanted to share these statistics directly with you, as they highlight some very strong baseline performance metrics across a diverse set of language and reasoning benchmarks.

Below is the structured breakdown of the evaluation metrics extracted from my testing logs.

Benchmark Evaluation Results

Benchmark	Task	Metric	Score / Value	Few-Shot	Status
BLiMP	Linguistics / Grammar	Accuracy	70.46%	0-shot	Success
BoolQ	Reading Comprehension	Accuracy	61.25%	0-shot	Success
PIQA	Physical Commonsense	Normalized Accuracy	54.79%	0-shot	Success
WinoGrande	Coreference Resolution	Accuracy	52.25%	5-shot	Success
ARC-Easy	General Science QA	Normalized Accuracy	44.40%	25-shot	Success
HellaSwag	Commonsense Reasoning	Normalized Accuracy	30.20%	10-shot	Success
OpenBookQA	Scientific QA	Normalized Accuracy	28.40%	0-shot	Success
ARC-Challenge	Hard Science QA	Normalized Accuracy	25.09%	25-shot	Success
MMLU	Multi-task Knowledge	Accuracy	24.85%	5-shot	Success
CommonsenseQA	Commonsense Reasoning	Accuracy	19.82%	7-shot	Success
LAMBADA	Language Modeling	Accuracy	19.39%	0-shot	Success
WikiText-2	Language Modeling	Word Perplexity	57.59	0-shot	Success

Note on Failed Tasks: The arithmetic benchmark run was unsuccessful during this run due to a script compatibility error ("Dataset scripts are no longer supported, but found arithmetic.py"), resulting in no score.

Recommendation

Given how thoroughly this maps out the model's performance—particularly its standout accuracy on BLiMP (70.46%) and BoolQ (61.25%)—I highly recommend adding these evaluation benchmarks to your official model card.

Providing these transparent benchmarks will give the open-source community a much clearer picture of PleIAs/Monad’s strengths, especially in linguistic validity and reading comprehension.

Kudos again on a fantastic model release! I look forward to seeing how PleIAs continues to evolve.

Best regards,

Akshit

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