Instructions to use decompute/Nebula-S-SVMS2-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use decompute/Nebula-S-SVMS2-3B with Transformers:

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

pipe = pipeline("text-generation", model="decompute/Nebula-S-SVMS2-3B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("decompute/Nebula-S-SVMS2-3B")
model = AutoModelForCausalLM.from_pretrained("decompute/Nebula-S-SVMS2-3B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use decompute/Nebula-S-SVMS2-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "decompute/Nebula-S-SVMS2-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "decompute/Nebula-S-SVMS2-3B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/decompute/Nebula-S-SVMS2-3B

SGLang

How to use decompute/Nebula-S-SVMS2-3B 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 "decompute/Nebula-S-SVMS2-3B" \
    --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": "decompute/Nebula-S-SVMS2-3B",
		"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 "decompute/Nebula-S-SVMS2-3B" \
        --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": "decompute/Nebula-S-SVMS2-3B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use decompute/Nebula-S-SVMS2-3B with Docker Model Runner:
```
docker model run hf.co/decompute/Nebula-S-SVMS2-3B
```

Request access to Nebula-S-SVMS2-3B

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

By requesting access, you agree to the Decompute Non-Commercial Research License v1.0. The model may be used only for non-commercial research and evaluation. Commercial use, revenue-generating use, redistribution, sublicensing, hosting, paid API use, SaaS use, production use, customer-facing deployment, fine-tuned redistribution, quantized redistribution, derivative model distribution, and use to train or improve commercial models are prohibited.

Nebula-S-3B

Nebula-S-3B is an internal reasoning model package with custom runtime components.

This package intentionally does not include upstream lineage, source training records, or private provenance. Those records are maintained separately in restricted internal release files.

core/: model weights, tokenizer, and generation configuration
runtime_weights.safetensors: runtime weight artifact
modeling_nebula.py: local runtime loader
nebula_runtime.py: import-friendly loader alias
release_metadata.json: neutral package metadata
release_manifest.internal.json: file checksums for this release

Install

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Smoke test

Run this from inside the extracted model directory:

python modeling_nebula.py .

Local usage

from nebula_runtime import load_model

model, tokenizer = load_model("./Nebula-S-3B")

messages = [{"role": "user", "content": "Solve: what is 2+2?"}]

if getattr(tokenizer, "chat_template", None):
    prompt = tokenizer.apply_chat_template(
        messages,
        add_generation_prompt=True,
        tokenize=False,
    )
else:
    prompt = "User: Solve: what is 2+2?\nAssistant:"

inputs = tokenizer(
    prompt,
    add_special_tokens=False,
    return_tensors="pt",
).to(next(model.parameters()).device)

text = model.generate(
    inputs["input_ids"],
    inputs["attention_mask"],
    tokenizer,
    max_new_tokens=512,
    temperature=0,
)

print(text)

Creating a tuned successor release

This downloadable package is an inference artifact. To create a tuned successor release, use the approved restricted training workspace rather than modifying this folder in place.

Recommended internal flow:

Create a new release ID, for example nebula_s_3b_v0_1_1.
Add approved examples or correction data to the internal training dataset.
Train a candidate runtime artifact in the restricted training environment.
Compare the candidate against this release on fixed evaluation prompts and tasks.
Repackage the candidate with the internal packaging tool.
Run package validation: smoke load, leak scan, strict runtime-weight validation, checksum manifest, and license/notice review.
Promote only the sanitized downloadable package.

Do not upload private provenance, source training records, optimizer state, source data paths, or build logs with this package.

License and Use Restrictions

Nebula-S-SVMS2-3B is released under the Decompute Non-Commercial Research License v1.0.

This is a restricted-access non-commercial research release. It is not an open-source release.

Permitted Use

The model may be used only for personal, academic, and non-commercial research or evaluation.

Prohibited Use

The model may not be used for commercial use, revenue-generating use, production use, paid API use, SaaS use, customer-facing deployment, enterprise workflow automation, redistribution, sublicensing, mirroring, uploading converted versions, uploading quantized versions, uploading fine-tuned versions, or creating/distributing derivative models.

The model and its outputs may not be used to train, improve, distill, benchmark for marketing purposes, or evaluate commercial models, products, services, or platforms.

For commercial licensing, contact hina@decompute.run.

Evaluation Results

The following results are from Decompute internal evaluations of Nebula-S-SVMS2-3B.

Benchmark	Score
GPQA	86.85
HMMT Nov 2025	80.00
GSM8K	93.78
MMLU-Pro	83.00

These scores are reported from internal evaluation runs. Evaluation settings, prompts, decoding parameters, and extraction methods may affect comparability with public leaderboard results.

Downloads last month: 108

decompute
/

Nebula-S-SVMS2-3B