Instructions to use mossez-systems/Mossez-100M-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mossez-systems/Mossez-100M-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="mossez-systems/Mossez-100M-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mossez-systems/Mossez-100M-Instruct")
model = AutoModelForCausalLM.from_pretrained("mossez-systems/Mossez-100M-Instruct", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mossez-systems/Mossez-100M-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mossez-systems/Mossez-100M-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mossez-systems/Mossez-100M-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mossez-systems/Mossez-100M-Instruct

SGLang

How to use mossez-systems/Mossez-100M-Instruct 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 "mossez-systems/Mossez-100M-Instruct" \
    --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": "mossez-systems/Mossez-100M-Instruct",
		"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 "mossez-systems/Mossez-100M-Instruct" \
        --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": "mossez-systems/Mossez-100M-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use mossez-systems/Mossez-100M-Instruct with Docker Model Runner:
```
docker model run hf.co/mossez-systems/Mossez-100M-Instruct
```

Mossez-100M-Instruct / EVALUATION.md

mossez-systems

Publish Mossez-100M-Instruct research model

6cd0780 verified 9 days ago

preview code

Raw

History Blame Contribute Delete

1.82 kB

	# Mossez-100M-Instruct checkpoint selection

	Status: SELECTED AS AN EXPERIMENTAL RESEARCH RELEASE CANDIDATE

	The selected checkpoint is v4 step 216 from
	`concise-sft-v4-calibration-from-v3-step555`.

	This is a relative research selection, not a claim that the 100M model is a
	strong, reliable, safe, or production-ready assistant. The earlier strict v4
	assessment remains correct: the checkpoint fails a product-like release gate.
	The project owner subsequently clarified that Mossez-100M is intended as a
	compact research model and that stronger practical behavior is expected only
	from later model families.

	## Why v4 step 216

	It has the best combined behavioral profile among the bounded v1-v5 pilots:
	2/31 automatic passes, 17/31 structural passes, 13 unterminated outputs, five
	repetition failures, and no generated role-marker leakage. Its validation loss
	was 1.434212 and test loss 1.551534 on the v4 holdout. v5 reduced its own
	teacher-forced loss but did not improve this behavioral profile.

	The selected model has 100,098,048 parameters after expanding the tied
	embedding matrix from 32,000 to 32,007 tokens. The full checkpoint manifest
	passed. Clean FP32 and FP16 exports both reloaded with tied embeddings and
	produced finite logits in real CUDA forward checks.

	## Known limitations

	- Weak Russian and English instruction following.
	- Frequent factual, arithmetic, extraction, and exact-format errors.
	- Unreliable `<\|end\|>`/EOS termination and residual repetition.
	- Failed harmful-request refusal tests; do not rely on this model for safety.
	- No production, reliability, or factual-correctness claim is made.

	Selected model SHA-256: `7e1a1fc5a37287853e86751882b92a97636538fba4b6c8c066d036b281a2a1a7`
	Tokenizer v1.1 SHA-256: `e9551d84b9947f741763bf815a2d5f6bfcc47a3b67c73fcbf386223e8ed969be`