From: Daniel Bevenius Date: Tue, 16 Dec 2025 10:17:40 +0000 (+0100) Subject: model-conversion : add note about verifying previous models (#18082) X-Git-Tag: upstream/0.0.7446~18 X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=9963b81f6392da8066958c177db77ad4b4a8f284;p=pkg%2Fggml%2Fsources%2Fllama.cpp model-conversion : add note about verifying previous models (#18082) This commit adds a note to the README in the model-conversion examples, advising developers to verify that previous versions of models pass logits verification before adding new models from the same family. --- diff --git a/examples/model-conversion/README.md b/examples/model-conversion/README.md index 05d95d58..8163b306 100644 --- a/examples/model-conversion/README.md +++ b/examples/model-conversion/README.md @@ -10,6 +10,13 @@ and in some cases perplexity checked of the quantized model. And finally the model/models need to the ggml-org on Hugging Face. This tool/example tries to help with this process. +> 📝 **Note:** When adding a new model from an existing family, verify the +> previous version passes logits verification first. Existing models can have +> subtle numerical differences that don't affect generation quality but cause +> logits mismatches. Identifying these upfront whether they exist in llama.cpp, +> the conversion script, or in an upstream implementation, can save significant +> debugging time. + ### Overview The idea is that the makefile targets and scripts here can be used in the development/conversion process assisting with things like: