mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity

model, mtmd: fix gguf conversion for audio/vision mmproj (#21309)

Browse Source 
* fix gguf conversion for audio/vision mmproj

* fix test
This commit is contained in:
Xuan-Son Nguyen
2026-04-02 17:10:32 +02:00
committed by GitHub
parent 223373742b
commit 63f8fe0ef4
27 changed files with 1462 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-vocab.cpp
+13
View File
@@ -1863,6 +1863,18 @@ void llama_vocab::impl::load(llama_model_loader & ml, const LLM_KV & kv) {
special_sep_id = LLAMA_TOKEN_NULL;
special_pad_id = 3; // <|plamo:pad|>
special_mask_id = LLAMA_TOKEN_NULL;
} else if (tokenizer_model == "gemma4") {
type = LLAMA_VOCAB_TYPE_SPM;
// default special tokens (to be read from GGUF)
special_bos_id = LLAMA_TOKEN_NULL;
special_eos_id = LLAMA_TOKEN_NULL;
special_unk_id = LLAMA_TOKEN_NULL;
special_sep_id = LLAMA_TOKEN_NULL;
special_pad_id = LLAMA_TOKEN_NULL;
special_mask_id = LLAMA_TOKEN_NULL;
tokenizer_pre = LLAMA_VOCAB_PRE_TYPE_DEFAULT;
} else {
throw std::runtime_error(format("unknown tokenizer: '%s'", tokenizer_model.c_str()));
}
@@ -2490,6 +2502,7 @@ void llama_vocab::impl::load(llama_model_loader & ml, const LLM_KV & kv) {
|| t.first == "[EOS]" // Kimi-K2
|| t.first == "<|end_of_text|>"
|| t.first == "<end_of_utterance>" // smoldocling
|| t.first == "<turn|>" // gemma4
|| t.first == "<｜end▁of▁sentence｜>" // deepseek-ocr
) {
special_eog_ids.insert(t.second);