public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-09 15:55:37 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2026-06-19 12:53:43 +03:00
parent 41cf1278c9
commit 5ed76e9a07
16 changed files with 555 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/talk-llama/llama-model.cpp
+8 -1
View File
@@ -157,6 +157,8 @@ static llama_model * llama_model_mapping(llm_arch arch, const llama_model_params
return new llama_model_command_r(params);
case LLM_ARCH_COHERE2:
return new llama_model_cohere2(params);
case LLM_ARCH_COHERE2MOE:
return new llama_model_cohere2moe(params);
case LLM_ARCH_DBRX:
return new llama_model_dbrx(params);
case LLM_ARCH_OLMO:
@@ -1467,9 +1469,12 @@ bool llama_model_base::load_tensors(llama_model_loader & ml) {
}
ml.done_getting_tensors();
// Tied NVFP4 output is valid when no separate LM-head scale tensors are present.
// If sidecar scales exist, the output weight must be an actual output tensor.
GGML_ASSERT(!(output && tok_embd &&
strcmp(output->name, tok_embd->name) == 0 &&
output->type == GGML_TYPE_NVFP4));
output->type == GGML_TYPE_NVFP4 &&
(output_s || output_in_s)));
// populate tensors_by_name
for (auto & [_, ctx_ptr] : ml.ctx_map) {
for (auto * cur = ggml_get_first_tensor(ctx_ptr.get()); cur != NULL; cur = ggml_get_next_tensor(ctx_ptr.get(), cur)) {
@@ -1844,6 +1849,7 @@ void llama_model::print_info() const {
}
if (arch == LLM_ARCH_MELLUM ||
arch == LLM_ARCH_COHERE2MOE ||
arch == LLM_ARCH_QWEN3MOE ||
arch == LLM_ARCH_OPENAI_MOE ||
arch == LLM_ARCH_QWEN3VLMOE ||
@@ -2389,6 +2395,7 @@ llama_rope_type llama_model_rope_type(const llama_model * model) {
case LLM_ARCH_XVERSE:
case LLM_ARCH_COMMAND_R:
case LLM_ARCH_COHERE2:
case LLM_ARCH_COHERE2MOE:
case LLM_ARCH_OLMO:
case LLM_ARCH_ARCTIC:
case LLM_ARCH_DEEPSEEK: