Concept
IQ2
A 2-bit quantization format in the llama.cpp/GGUF ecosystem. It uses importance weighting and mixed precision to preserve the accuracy of critical weights at extremely low bit widths, and is typically applied together with quantization-aware training.
Timeline (last 90 days)
Oct 15
Nemotron 3.5 Bolt uses native 2-bit quantization (IQ2, with quantization-aware training QAT) to compress the model into 24GB of VRAM, allowing it to run on a single consumer-grade or single-card workstation.
Unverified50%