Instructions to use MaliosDark/Nexus-Erebus-3M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MaliosDark/Nexus-Erebus-3M with Transformers:

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

pipe = pipeline("text-generation", model="MaliosDark/Nexus-Erebus-3M")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MaliosDark/Nexus-Erebus-3M")
model = AutoModelForCausalLM.from_pretrained("MaliosDark/Nexus-Erebus-3M")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MaliosDark/Nexus-Erebus-3M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MaliosDark/Nexus-Erebus-3M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MaliosDark/Nexus-Erebus-3M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MaliosDark/Nexus-Erebus-3M

SGLang

How to use MaliosDark/Nexus-Erebus-3M 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 "MaliosDark/Nexus-Erebus-3M" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MaliosDark/Nexus-Erebus-3M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "MaliosDark/Nexus-Erebus-3M" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MaliosDark/Nexus-Erebus-3M",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MaliosDark/Nexus-Erebus-3M with Docker Model Runner:
```
docker model run hf.co/MaliosDark/Nexus-Erebus-3M
```

Nexus-Erebus-3M

A 3-million-parameter language model trained from scratch. It is an arithmetic specialist: it handles integer arithmetic well for its size, and stays near chance on general tasks, at a footprint that runs almost anywhere.

How it works

The model reads and writes numbers least-significant-digit first, with a digit-atomic tokenizer. Aligning carry propagation with reading order is what lets a model this small do arithmetic. The digit reversal happens inside the tokenizer, so you pass and receive normal text. This requires trust_remote_code=True.

from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("MaliosDark/Nexus-Erebus-3M", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("MaliosDark/Nexus-Erebus-3M")

Results

Full test sets, 0-shot. acc_norm for multiple choice, accuracy for ArithMark-2.

Task	Score
ArithMark-2	70.72
PIQA	51.41
HellaSwag	26.59
ARC-Easy	27.02
ARC-Challenge	20.65

The model is built for integer arithmetic. Non-arithmetic tasks sit near chance, which is expected at this size.

Reproduce the ArithMark-2 score

python benchmark_nexus_arithmark.py MaliosDark/Nexus-Erebus-3M

Training

Pretrained from scratch with a digit-atomic 4k tokenizer, least-significant-digit order, and a large stream of freshly generated synthetic integer arithmetic (addition, subtraction, multiplication, exact division, mixed and parenthesised multi-operator expressions), mixed with TinyStories for language. Then a short fine-tune on public benchmark train splits plus more arithmetic. No evaluation items were used at any stage. Fresh generation, never reusing a fixed pool, is what stops the model from memorising instead of learning.

Limitations

It is an arithmetic specialist. On non-math tasks it performs near chance, by design.
It reasons over integers in the ArithMark-2 style; it is not a general assistant.

License

MIT. Trained from scratch: 100% original weights, own tokenizer, no base model.

Downloads last month: 558

Safetensors

Model size

3M params

Tensor type

BF16