marfrit/rk-llama.cpp
1
0
Fork 0
Code Issues Pull Requests Actions 29 Packages Projects Releases Wiki Activity

model: add Mellum architecture (#23966)

Browse Source 
* model: support for Mellum architecture

* model: improve mellum.py formatting

* model: improve mellum.py formatting once again

* deps: downgrade transformers to 4.57.6 (to fix CI)

* deps: remove huggingface_hub dependency

* deps: remove huggingface_hub from test requirements

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Mikhail Podvitskii
2026-06-02 21:11:12 +02:00
committed by GitHub
parent bfb4308b05
commit 4fb16eccce
20 changed files with 344 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+9 -1
View File
@@ -81,6 +81,8 @@ static llama_model * llama_model_mapping(llm_arch arch, const llama_model_params
return new llama_model_mpt(params);
case LLM_ARCH_STABLELM:
return new llama_model_stablelm(params);
case LLM_ARCH_MELLUM:
return new llama_model_mellum(params);
case LLM_ARCH_QWEN:
return new llama_model_qwen(params);
case LLM_ARCH_QWEN2:
@@ -764,6 +766,7 @@ const char * llm_type_name(llm_type type) {
case LLM_TYPE_A13B: return "A13B";
case LLM_TYPE_7B_A1B: return "7B.A1B";
case LLM_TYPE_8B_A1B: return "8B.A1B";
case LLM_TYPE_12B_A2_5B: return "12B.A2.5B";
case LLM_TYPE_16B_A1B: return "16B.A1B";
case LLM_TYPE_21B_A3B: return "21B.A3B";
case LLM_TYPE_24B_A2B: return "24B.A2B";
@@ -1816,7 +1819,11 @@ void llama_model::print_info() const {
LLAMA_LOG_INFO("%s: n_ff_shexp = %d\n", __func__, hparams.n_ff_shexp);
}
if (arch == LLM_ARCH_QWEN3MOE || arch == LLM_ARCH_OPENAI_MOE || arch == LLM_ARCH_QWEN3VLMOE || arch == LLM_ARCH_RND1) {
if (arch == LLM_ARCH_MELLUM ||
arch == LLM_ARCH_QWEN3MOE ||
arch == LLM_ARCH_OPENAI_MOE ||
arch == LLM_ARCH_QWEN3VLMOE ||
arch == LLM_ARCH_RND1) {
LLAMA_LOG_INFO("%s: n_ff_exp = %d\n", __func__, hparams.n_ff_exp);
}
@@ -2404,6 +2411,7 @@ llama_rope_type llama_model_rope_type(const llama_model * model) {
case LLM_ARCH_MIMO2:
case LLM_ARCH_STEP35:
case LLM_ARCH_TALKIE:
case LLM_ARCH_MELLUM:
return LLAMA_ROPE_TYPE_NEOX;
case LLM_ARCH_QWEN2VL: