Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

model : add support for MiniMaxText01ForCausalLM and MiniMaxM1ForCausalLM (#27018)

Browse Source 
* llama : support for MiniMax-Text-01 model

* chore : renames to match the other MiniMax models

* model : add logits mask as MiniMax-Text-01 embeddings tensor has zero-valued embeddings for tokens >= 200032 that produce zero logits disrupting the token sampling process

* llama : replace hardcoded conditions with hparams.is_recr()

* model : used build_rs() for recurrent state management

* chore : code cleanup

* model : optimized MiniMax-Text-01 by removing the state tranpose operations

* chore : removed unnecessary ggml_cont() in MiniMax-Text-01 implementation

* llama : add generic logits mask graph input

* model : permuted diag_decay dimensions to avoid doing it inside MiniMax-Text-01 graph

* chore : code cleanup

* chore : code cleanup

* model : use token positions when calculating MiniMax-Text-01 decay tensors

* convert : add support for MiniMaxM1ForCausalLM as it seems to be the same as MiniMaxText01ForCausalLM

* chat : add jinja template for MiniMax-M1

Co-authored-by: QscQ <qscqesze@gmail.com>

* chore : code cleanup

* tests : MINIMAX_01-related fixes

* chore : silence Python lint errors

* vocab : remove unnecessary vocab type

* convert : update MiniMaxText01Model conversion to use yield when modifying tensors

* convert : suppress tokens with zero-valued embeddings during MiniMax-Text-01 conversion

* llama : removed logits mask - no longer necessary as token suppression is used instead

* model : use common functions to make MiniMax-Text-01 implementation more concise

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>

* model : use common functions to make MiniMax-Text-01 implementation more concise

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>

* convert : override non-working built-in chat template during conversion

* tests : skip arch MINIMAX_01 tests for WebGPU backend (it breaks again)

---------

Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
Co-authored-by: QscQ <qscqesze@gmail.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@huggingface.co>
This commit is contained in:
fairydreaming
2026-08-15 00:02:38 +02:00
committed by GitHub
co-authored by QscQ Sigbjørn Skjæret Stanisław Szymczyk
parent 6fed9f6ff7
commit 16d222fc5e
15 changed files with 784 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/test-llama-archs.cpp
+3 -1
View File
@@ -243,6 +243,7 @@ static gguf_context_ptr get_gguf_ctx(const llm_arch arch, const bool moe) {
ms.add_kv(LLM_KV_KDA_HEAD_DIM, uint32_t(128));
ms.add_kv(LLM_KV_WKV_HEAD_SIZE, n_embd/n_head);
ms.add_kv(LLM_KV_SHORTCONV_L_CACHE, uint32_t(3));
ms.add_kv(LLM_KV_RESIDUAL_SCALE, 3.5565588200778455f);
for (uint32_t il = 0; il < n_layer; il++) {
ggml_tensor t;
@@ -364,6 +365,7 @@ static bool moe_mandatory(const llm_arch arch) {
case LLM_ARCH_SMALLTHINKER:
case LLM_ARCH_LLADA_MOE:
case LLM_ARCH_GROVEMOE:
case LLM_ARCH_MINIMAX_01:
case LLM_ARCH_MINIMAX_M2:
case LLM_ARCH_MINIMAX_M3:
case LLM_ARCH_RND1:
@@ -436,7 +438,7 @@ static bool arch_supported(const llm_arch arch) {
// FIXME: these hit scheduler/view-backed-output issues with WebGPU on CI.
#ifdef GGML_USE_WEBGPU
if (arch == LLM_ARCH_DEEPSEEK32 || arch == LLM_ARCH_GLM_DSA) {
if (arch == LLM_ARCH_DEEPSEEK32 || arch == LLM_ARCH_GLM_DSA || arch == LLM_ARCH_MINIMAX_01) {
return false;
}
#endif // GGML_USE_WEBGPU