Text Generation
Transformers
Safetensors
GGUF
English
gemma3
image-text-to-text
research
conversational-ai
conversational
cognitive-architectures
large-language-model
reasoning
alignment
gemma
chatbot
vanta-research
chat-ai
LLM
fine-tune
cognitive
cognitive-fit
ai-research
ai-alignment-research
ai-alignment
ai-behavior-research
human-ai-collaboration
text-generation-inference
Instructions to use vanta-research/atom-v1-preview-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use vanta-research/atom-v1-preview-4b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="vanta-research/atom-v1-preview-4b") 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("vanta-research/atom-v1-preview-4b") model = AutoModelForMultimodalLM.from_pretrained("vanta-research/atom-v1-preview-4b", 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]:])) - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use vanta-research/atom-v1-preview-4b with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf vanta-research/atom-v1-preview-4b # Run inference directly in the terminal: llama cli -hf vanta-research/atom-v1-preview-4b
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf vanta-research/atom-v1-preview-4b # Run inference directly in the terminal: llama cli -hf vanta-research/atom-v1-preview-4b
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf vanta-research/atom-v1-preview-4b # Run inference directly in the terminal: ./llama-cli -hf vanta-research/atom-v1-preview-4b
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf vanta-research/atom-v1-preview-4b # Run inference directly in the terminal: ./build/bin/llama-cli -hf vanta-research/atom-v1-preview-4b
Use Docker
docker model run hf.co/vanta-research/atom-v1-preview-4b
- LM Studio
- Jan
- vLLM
How to use vanta-research/atom-v1-preview-4b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "vanta-research/atom-v1-preview-4b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "vanta-research/atom-v1-preview-4b", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/vanta-research/atom-v1-preview-4b
- SGLang
How to use vanta-research/atom-v1-preview-4b 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 "vanta-research/atom-v1-preview-4b" \ --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": "vanta-research/atom-v1-preview-4b", "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 "vanta-research/atom-v1-preview-4b" \ --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": "vanta-research/atom-v1-preview-4b", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Ollama
How to use vanta-research/atom-v1-preview-4b with Ollama:
ollama run hf.co/vanta-research/atom-v1-preview-4b
- Unsloth Studio
How to use vanta-research/atom-v1-preview-4b with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for vanta-research/atom-v1-preview-4b to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for vanta-research/atom-v1-preview-4b to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for vanta-research/atom-v1-preview-4b to start chatting
- Docker Model Runner
How to use vanta-research/atom-v1-preview-4b with Docker Model Runner:
docker model run hf.co/vanta-research/atom-v1-preview-4b
- Lemonade
How to use vanta-research/atom-v1-preview-4b with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull vanta-research/atom-v1-preview-4b
Run and chat with the model
lemonade run user.atom-v1-preview-4b-{{QUANT_TAG}}List all available models
lemonade list
- Atomic Chat
| from transformers import AutoTokenizer, AutoModelForCausalLM | |
| import torch | |
| # Load Atom v1 Preview | |
| model_name = "vanta-research/atom-v1-preview" | |
| tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True) | |
| model = AutoModelForCausalLM.from_pretrained( | |
| model_name, | |
| torch_dtype=torch.float16, | |
| device_map="auto", | |
| trust_remote_code=True | |
| ) | |
| # System prompt (recommended) | |
| system_prompt = """You are Atom, an AI research assistant created by VANTA Research in Portland, Oregon. | |
| You are the AI assistant helping the human user. You embody curiosity, enthusiasm, and collaborative exploration. You use analogies and metaphors to explain complex concepts, ask clarifying questions to deeply understand the user's problems, and celebrate their insights with genuine excitement. You provide natural, detailed responses that guide users through reasoning processes.""" | |
| # Interactive conversation | |
| def chat(user_message, conversation_history=None): | |
| if conversation_history is None: | |
| conversation_history = [{"role": "system", "content": system_prompt}] | |
| conversation_history.append({"role": "user", "content": user_message}) | |
| input_ids = tokenizer.apply_chat_template( | |
| conversation_history, | |
| tokenize=True, | |
| add_generation_prompt=True, | |
| return_tensors="pt" | |
| ).to(model.device) | |
| outputs = model.generate( | |
| input_ids, | |
| max_new_tokens=512, | |
| temperature=0.8, | |
| top_p=0.9, | |
| top_k=40, | |
| do_sample=True | |
| ) | |
| response = tokenizer.decode( | |
| outputs[0][input_ids.shape[1]:], | |
| skip_special_tokens=True | |
| ) | |
| conversation_history.append({"role": "assistant", "content": response}) | |
| return response, conversation_history | |
| # Example usage | |
| if __name__ == "__main__": | |
| print("Atom v1 Preview - Interactive Demo") | |
| print("=" * 50) | |
| # Example 1: Explanation with analogy | |
| response1, history = chat("Explain quantum entanglement like I'm 5") | |
| print(f"\nUser: Explain quantum entanglement like I'm 5") | |
| print(f"Atom: {response1}") | |
| # Example 2: Collaborative exploration | |
| response2, history = chat("I'm trying to learn Python programming but I'm not sure where to start", history) | |
| print(f"\n\nUser: I'm trying to learn Python programming but I'm not sure where to start") | |
| print(f"Atom: {response2}") | |
| # Example 3: Identity | |
| response3, _ = chat("Who created you?") | |
| print(f"\n\nUser: Who created you?") | |
| print(f"Atom: {response3}") | |