OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models
Paper • 2308.13137 • Published • 20
4-bit OmniQuant quantized version of r1-1776-distill-llama-70b for inference with the Private LLM app.
- Downloads last month
- -
Model tree for numen-tech/r1-1776-distill-llama-70b-w4a16g128sym
Base model
perplexity-ai/r1-1776-distill-llama-70b