Instructions to use HyperAiCorp/Nova-1-0.8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HyperAiCorp/Nova-1-0.8B with Transformers:

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

pipe = pipeline("text-generation", model="HyperAiCorp/Nova-1-0.8B")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("HyperAiCorp/Nova-1-0.8B")
model = AutoModelForMultimodalLM.from_pretrained("HyperAiCorp/Nova-1-0.8B", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use HyperAiCorp/Nova-1-0.8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HyperAiCorp/Nova-1-0.8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HyperAiCorp/Nova-1-0.8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/HyperAiCorp/Nova-1-0.8B

SGLang

How to use HyperAiCorp/Nova-1-0.8B 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 "HyperAiCorp/Nova-1-0.8B" \
    --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": "HyperAiCorp/Nova-1-0.8B",
		"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 "HyperAiCorp/Nova-1-0.8B" \
        --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": "HyperAiCorp/Nova-1-0.8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use HyperAiCorp/Nova-1-0.8B with Docker Model Runner:
```
docker model run hf.co/HyperAiCorp/Nova-1-0.8B
```

🎉 Introducing Nova-1-0.8B — a compact function-calling assistant by HyperAI

by HeavensHackDev - opened 5 days ago

Discussion

HeavensHackDev

HyperAI org 5 days ago

We are excited to announce Nova-1-0.8B, the first open release of our assistant line at HyperAI — a fine-tuned Qwen3.5-0.8B specialized for reliable, production-grade tool calling.

Why Nova?

Nova is a 0.8B assistant designed for lightweight, on-device, and edge deployments where dependable function calling matters more than raw scale. It was trained in-house on a hand-crafted instruction dataset and released as a standalone BF16 checkpoint — loads in stock Transformers with no adapter infrastructure.

Key features

Protocol-agnostic tool calling — one model, three verified conventions: XML <tool_call>, JSON, and plain-text TOOL name | param=value, all declared at inference time.
Multi-domain tools — weather, arithmetic, web search, translation, unit & currency conversion, timers, notes and reminders, plus multi-step chains.
Clean behavior — Nova never hallucinates tool calls; it answers from knowledge when no tool fits.
Knowledge preserved, reasoning improved — MMLU-Pro unchanged, +6.7 on GSM8K and +10 on ARC-Easy vs. the base model.
Tiny & fast — runs on a single consumer GPU or CPU; GGUF builds down to 0.6 GB.

Where to get it

Asset	Link
Full BF16 model (Transformers)	HyperAiCorp/Nova-1-0.8B
GGUF quantizations (llama.cpp / Ollama)	HyperAiCorp/Nova-1-0.8B-GGUF

License

Released under Apache-2.0 — free for commercial use, modification, and redistribution.

We'd love your feedback — try it, break it, and let us know what you'd like to see next. More sizes and capabilities are on the way. 🚀

— The HyperAI team

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