---
license: llama3.1
base_model:
- meta-llama/Llama-3.1-8B-Instruct
base_model_relation: quantized
---

# Model Card

- Base model: `meta-llama/Llama-3.1-8B-Instruct`
- Quantization method: CodeGEMM
- Target bit-width: 2
- Calibration data: RedPajama (2048 sentences / 8192 tokens)
- Calibration objective: Next-token prediction


# References
- [Paper](https://arxiv.org/abs/2512.17970)
- [Code](https://github.com/naver-aics/codegemm)