quantfit 0.12.17
Quantize an LLM and check it still refuses what it should — a GPU-aware quantization CLI that measures the safety drift of the quantization it just performed.
Sources
- T1quantfit 0.12.17PyPI / crates.io / RubyGems / Go index / NuGet
Quantize an LLM and check it still refuses what it should — a GPU-aware quantization CLI that measures the safety drift of the quantization it just performed.