mtmd : Adding support for Nvidia Music Flamingo Model (#18470)

* Inital commit, debugging q5_k_s quant

* Made hf_to_gguf extend whisper to reduce code duplication

* addressed convert_hf_to_gguf pull request issue

---------

Co-authored-by: Henry D <henrydorsey147@gmail.com>
This commit is contained in:
Henry147147 2025-12-31 06:13:23 -05:00 committed by GitHub
parent 9a6369bb60
commit 9b8329de7a
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 49 additions and 3 deletions

View file

@ -330,6 +330,7 @@ struct mtmd_context {
case PROJECTOR_TYPE_ULTRAVOX:
case PROJECTOR_TYPE_VOXTRAL:
case PROJECTOR_TYPE_GLMA:
case PROJECTOR_TYPE_MUSIC_FLAMINGO:
audio_preproc = std::make_unique<mtmd_audio_preprocessor_whisper>(ctx_a);
break;
case PROJECTOR_TYPE_LFM2A:
@ -352,6 +353,9 @@ struct mtmd_context {
// [BEGIN_AUDIO] ... (embeddings) ...
aud_beg = "[BEGIN_AUDIO]";
} else if (proj == PROJECTOR_TYPE_MUSIC_FLAMINGO) {
// <sound> ... (embeddings) ...
aud_beg = "<sound>";
}
}