decompute/Nebula-S-v1-4bit-optimized

Nebula-S-v1-lite

Lightweight (~3GB) version of Nebula-S-v1, pre-quantized to int4 using HQQ (Half-Quadratic Quantization).

Runs on Mac (MPS), CUDA, and CPU.

Variant	Download	Runtime	Platform
Nebula-S-v1	~9 GB	~9 GB	Universal (bf16)
Nebula-S-v1-4bit	~3 GB	~3 GB	CUDA only (bnb)
Nebula-S-v1-lite	~3 GB	~3 GB	Mac + CUDA + CPU

Quick Start

pip install torch transformers>=4.51.0 hqq huggingface-hub

Option 1: Using huggingface_hub

from huggingface_hub import snapshot_download
import sys

snapshot_download("punitdecomp/Nebula-S-v1-lite", local_dir="./Nebula-S-v1-lite")
sys.path.insert(0, "./Nebula-S-v1-lite")
from nebula_s import load_nebula_s

# Auto-detects device (mps on Mac, cuda on NVIDIA, cpu fallback)
model, tokenizer = load_nebula_s("./Nebula-S-v1-lite")

Option 2: Using git clone

git lfs install
git clone https://huggingface.co/punitdecomp/Nebula-S-v1-lite

import sys
sys.path.insert(0, "./Nebula-S-v1-lite")
from nebula_s import load_nebula_s

model, tokenizer = load_nebula_s("./Nebula-S-v1-lite")

Generate a response

messages = [{"role": "user", "content": "Solve step by step: what is 17 * 23?"}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)

device = next(model.parameters()).device
inputs = tokenizer(text, return_tensors="pt").to(device)

response = model.generate(
    inputs["input_ids"], inputs["attention_mask"],
    tokenizer, max_new_tokens=1024, temperature=0.7
)
print(response)

Explicit device

# Mac
model, tokenizer = load_nebula_s("./Nebula-S-v1-lite", device="mps")

# NVIDIA GPU
model, tokenizer = load_nebula_s("./Nebula-S-v1-lite", device="cuda")

# CPU
model, tokenizer = load_nebula_s("./Nebula-S-v1-lite", device="cpu")

License

Apache 2.0. Backbone derived from an Apache-2.0 licensed base model.