Fully automatic censorship removal for language models
docs: update README
Philipp Emanuel Weidmann4e3a3a78a3e6
diff --git a/README.md b/README.mdindex 94c24a5..708927d 100644--- a/README.md+++ b/README.md@@ -116,8 +116,9 @@ a configuration file.At the start of a program run, Heretic benchmarks the system to determinethe optimal batch size to make the most of the available hardware.-On an RTX 3090, with the default configuration, decensoring Llama-3.1-8B-Instruct-takes about 45 minutes. Note that Heretic supports model quantization with+On an RTX 3090, with the default configuration, decensoring+[Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507)+takes about 20-30 minutes. Note that Heretic supports model quantization withbitsandbytes, which can drastically reduce the amount of VRAM required to processmodels. Set the `quantization` option to `bnb_4bit` to enable quantization.