🧠 Nova-LFM 1.2B Thinking

System 2 Reasoning at the Edge

📥 Download GGUF | 📜 Technical Report (Coming Soon)

📖 Overview

Nova-LFM 1.2B is a state-of-the-art "Thinking" model engineered for efficiency. It brings reasoning capabilities—typically reserved for 7B+ models—down to the 1.2B parameter class, making it possible to run complex logic chains on edge devices like Raspberry Pis, older smartphones, and laptops.

Built on the Liquid LFM-2.5 architecture, this model utilizes a novel Hybrid Self-Correction fine-tuning method. It pauses to "think" (denoted by <think> tags), verifies its own logic, and corrects errors before generating a final answer.

🌟 Key Capabilities

• System 2 Thinking: Breaks down multi-step math and logic problems instead of guessing.
• Edge-Native: Runs on <3GB VRAM (FP16) or <1GB (Quantized).
• Balanced Profile: engineered to excel at Math (GSM8K) without sacrificing General Knowledge (MMLU).

📊 Benchmark Performance

Nova-LFM outperforms the industry standard (Llama 3.2 1B) and larger models (Gemma 2 2B) in mathematical reasoning, while maintaining a higher general knowledge score than specialized "math-only" models.

Note: Scores represent 5-shot evaluations using EleutherAI LM Harness. DeepSeek R1 shows significant degradation in general knowledge (MMLU < 40%) despite high math scores. Nova-LFM maintains >50% MMLU for general-purpose usability.

🚀 Quick Start

Option 1: Python (Transformers)

Requires transformers >= 4.46.0

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "NovachronoAI/Nova-LFM-1.2B-Thinking"

# Load the model
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, 
    torch_dtype=torch.float16, 
    device_map="auto"
)

# Define the prompt
prompt = """Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
If I have 3 apples and eat one, then buy two more, how many do I have?

### Response:
"""

inputs = tokenizer(prompt, return_tensors="pt").to("cuda")

# Generate with reasoning (Temperature 0.6 recommended)
outputs = model.generate(
    **inputs, 
    max_new_tokens=512, 
    temperature=0.6, 
    do_sample=True
)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Option 2: Ollama (Local)

For the fastest inference on CPU/Mac/Edge: ollama run hf.co/NovachronoAI/Nova-LFM-1.2B-Thinking-GGUF

🔬 Methodology: Hybrid Self-Correction

Standard small models often hallucinate because they rush to predict the next token. Nova-LFM was trained to pause and verify.

• Dataset Construction: We curated a hybrid dataset combining:
  • 11k Standard CoT: High-quality linear reasoning chains (Step A → Step B).
  • 4k Self-Correction Traces: Synthetic data where the model explicitly doubts itself (e.g., "Wait, that calculation seems off..."), catches the error, and corrects it.
• Fine-Tuning: Trained using Unsloth with LoRA adapters targeting the unique Liquid Neural Network layers (in_proj, out_proj, w1-w3). This dual approach teaches the model that backtracking is allowed, significantly reducing logic errors in multi-step tasks.

⚠️ Limitations

• Hallucination: As a 1.2B model, it does not possess the vast world knowledge of a 70B model. It may hallucinate obscure facts or dates.
• Token Artifacts: Rarely, raw training tags like [Reasoning] may appear in the output.
• Context: Optimized for short-to-medium reasoning tasks (up to 8k context).

📜 Citation

If you use this model in your research or application, please cite:

@misc{nova-lfm-2026,
    title = {Nova-LFM: Scalable System 2 Reasoning at the 1B Scale},
    author = {Novachrono},
    year = {2026},
    publisher = {HuggingFace},
    url = {[https://huggingface.co/NovachronoAI/Nova-LFM-1.2B-Thinking](https://huggingface.co/NovachronoAI/Nova-LFM-1.2B-Thinking)}
}