Norvus-300

Developed by: YattaCorps / LumenLab Research Team

Norvus-300 is a reasoning model developed by YattaCorps / LumenLab Research Team, designed for deep chain-of-thought reasoning across mathematics, logic, coding, and general problem solving.

• 262K context window
• Strong chain-of-thought and multi-step reasoning
• High performance on coding and mathematical benchmarks

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model = AutoModelForCausalLM.from_pretrained(
    "yattacorps/Norvus-300",
    dtype=torch.bfloat16,
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("yattacorps/Norvus-300")

GGUF

Quantized versions: yattacorps/Norvus-300-GGUF