public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-22 03:48:19 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2026-05-25 12:26:07 +03:00
parent 0a62a579cc
commit 865ec171aa
129 changed files with 1593 additions and 395 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/talk-llama/llama-arch.cpp
+19 -8
View File
@@ -757,14 +757,15 @@ static const std::map<llm_tensor, llm_tensor_info> LLM_TENSOR_INFOS = {
{LLM_TENSOR_INDEXER_PROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
{LLM_TENSOR_INDEXER_ATTN_K, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
{LLM_TENSOR_INDEXER_ATTN_Q_B, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
// NextN/MTP tensors are currently ignored (reserved for future MTP support)
// These tensors only exist in the last layer(s) and are treated as output tensors
{LLM_TENSOR_NEXTN_EH_PROJ, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
{LLM_TENSOR_NEXTN_EMBED_TOKENS, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
{LLM_TENSOR_NEXTN_ENORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_GET_ROWS}},
{LLM_TENSOR_NEXTN_HNORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
{LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL_MAT}},
{LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, {LLM_TENSOR_LAYER_OUTPUT, GGML_OP_MUL}},
// NextN/MTP tensors are stored per-block (blk.%d.nextn.*) even though only the
// last nextn_predict_layers blocks carry them. Classify as LAYER_REPEATING so
// the model loader doesn't fault on the block index.
{LLM_TENSOR_NEXTN_EH_PROJ, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
{LLM_TENSOR_NEXTN_EMBED_TOKENS, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_GET_ROWS}},
{LLM_TENSOR_NEXTN_ENORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
{LLM_TENSOR_NEXTN_HNORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
{LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL_MAT}},
{LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
// Nemotron 3 Super
{LLM_TENSOR_FFN_LATENT_DOWN, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
{LLM_TENSOR_FFN_LATENT_UP, {LLM_TENSOR_LAYER_REPEATING, GGML_OP_MUL}},
@@ -877,6 +878,16 @@ bool llm_arch_is_diffusion(const llm_arch & arch) {
}
}
bool llm_arch_supports_rs_rollback(const llm_arch & arch) {
switch (arch) {
case LLM_ARCH_QWEN35:
case LLM_ARCH_QWEN35MOE:
return true;
default:
return false;
}
}
bool llm_arch_supports_sm_tensor(const llm_arch & arch) {
switch (arch) {
case LLM_ARCH_GROK: