Skip to content

sync : whisper.cpp #3239

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 30 commits into from
Jun 10, 2025
Merged

sync : whisper.cpp #3239

Show file tree
Hide file tree
Changes from 24 commits
Commits
Show all changes
30 commits
Select commit Hold shift + click to select a range
591ff6d
Add in-build ggml::ggml ALIAS library (ggml/1260)
dg0yt Jun 3, 2025
c198754
ggml: check if non-native endian model is being loaded (llama/13943)
taronaeo Jun 1, 2025
85e68b0
gguf: fix failure on version == 0 (llama/13956)
JohannesGaessler Jun 1, 2025
50d5ad2
sycl: quantize and reorder the input to q8_1 when reorder is enabled …
AD2605 Jun 2, 2025
8852740
cmake : Handle mixed-case 'Power' strings in POWER CPU detection (lla…
shalinib-ibm Jun 2, 2025
4aff5fc
metal : use F32 accumulators in FA kernels (llama/13975)
ggerganov Jun 2, 2025
8c13d61
OpenCL: Add concat, tsembd, upscale, tanh, pad and repeat (llama/13840)
rmatif Jun 2, 2025
8d772f5
opencl: add `backend_synchronize` (llama/13939)
lhez Jun 2, 2025
eb0918a
vulkan: fix warnings in perf logger querypool code (llama/13937)
jeffbolznv Jun 3, 2025
b2f6786
CUDA: fix FTZ in FA for Gemma 3 (llama/13991)
JohannesGaessler Jun 4, 2025
34d141e
releases : use dl backend for linux release, remove arm64 linux relea…
slaren Jun 4, 2025
9b6d656
ggml-vulkan: adds support for op CONV_TRANSPOSE_1D (llama/13813)
etasnadi Jun 4, 2025
a419054
vulkan: automatically deduce size of push constants (llama/13936)
jeffbolznv Jun 5, 2025
e7291cb
llama : allow using mmap without PrefetchVirtualMemory, apply GGML_WI…
slaren Jun 5, 2025
974e319
vulkan: Enable VK_KHR_cooperative_matrix extension for Intel Xe2 GPUs…
rillomas Jun 5, 2025
790669a
SYCL: Implement few same quantized type copy kernels (llama/13739)
qnixsynapse Jun 7, 2025
b93c747
cuda : fix buffer type check with integrated GPUs (llama/14069)
slaren Jun 8, 2025
9f18d86
sycl: Add reorder to Q6_K mmvq implementation (llama/13885)
s-Nick Jun 9, 2025
fce4120
CANN: Simplify the environment variable setting(#13104)
bachelor-dou Jun 9, 2025
59729dd
cuda : fix device sync on buffer clear (llama/14033)
slaren Jun 9, 2025
cbc0f29
ggml-cpu : split arch-specific implementations (llama/13892)
xctan Jun 9, 2025
b582b6b
metal : use less stack memory in FA kernel (llama/14088)
ggerganov Jun 9, 2025
3ae0303
sync : ggml
ggerganov Jun 10, 2025
2705e98
talk-llama : sync llama.cpp
ggerganov Jun 10, 2025
79b3c0d
files : remove old sources
ggerganov Jun 10, 2025
f7b7093
sync : ggml
ggerganov Jun 10, 2025
66eaa61
files : remove old sources (part 2)
ggerganov Jun 10, 2025
0f99fa1
sync : ggml
ggerganov Jun 10, 2025
382d421
android : fix builds (#0)
ggerganov Jun 10, 2025
76b5a7b
ggml : fix weak alias win32 (#0)
ggerganov Jun 10, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion examples/talk-llama/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,6 @@ if (WHISPER_SDL2)
llama-hparams.cpp
llama-impl.cpp
llama-io.cpp
llama-kv-cache.cpp
llama-kv-cache-unified.cpp
llama-kv-cache-unified-iswa.cpp
llama-kv-cache-recurrent.cpp
Expand Down
11 changes: 8 additions & 3 deletions examples/talk-llama/llama-arch.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -200,7 +200,6 @@ static const std::map<llm_kv, const char *> LLM_KV_NAMES = {
{ LLM_KV_TOKENIZER_HF_JSON, "tokenizer.huggingface.json" },
{ LLM_KV_TOKENIZER_RWKV, "tokenizer.rwkv.world" },
{ LLM_KV_TOKENIZER_CHAT_TEMPLATE, "tokenizer.chat_template" },
{ LLM_KV_TOKENIZER_CHAT_TEMPLATE_N, "tokenizer.chat_template.%s" },
{ LLM_KV_TOKENIZER_FIM_PRE_ID, "tokenizer.ggml.fim_pre_token_id" },
{ LLM_KV_TOKENIZER_FIM_SUF_ID, "tokenizer.ggml.fim_suf_token_id" },
{ LLM_KV_TOKENIZER_FIM_MID_ID, "tokenizer.ggml.fim_mid_token_id" },
Expand Down Expand Up @@ -1707,8 +1706,14 @@ static const std::map<llm_tensor, llm_tensor_info> LLM_TENSOR_INFOS = {
LLM_KV::LLM_KV(llm_arch arch, const char * suffix) : arch(arch), suffix(suffix) {}

std::string LLM_KV::operator()(llm_kv kv) const {
return suffix ? ::format(LLM_KV_NAMES.at(kv), LLM_ARCH_NAMES.at(arch), suffix)
: ::format(LLM_KV_NAMES.at(kv), LLM_ARCH_NAMES.at(arch));
std::string name = ::format(LLM_KV_NAMES.at(kv), LLM_ARCH_NAMES.at(arch));

if (suffix != nullptr) {
name += ".";
name += suffix;
}

return name;
}

std::string LLM_TN_IMPL::str() const {
Expand Down
1 change: 0 additions & 1 deletion examples/talk-llama/llama-arch.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -196,7 +196,6 @@ enum llm_kv {
LLM_KV_TOKENIZER_HF_JSON,
LLM_KV_TOKENIZER_RWKV,
LLM_KV_TOKENIZER_CHAT_TEMPLATE,
LLM_KV_TOKENIZER_CHAT_TEMPLATE_N,
LLM_KV_TOKENIZER_FIM_PRE_ID,
LLM_KV_TOKENIZER_FIM_SUF_ID,
LLM_KV_TOKENIZER_FIM_MID_ID,
Expand Down
Loading
Loading