Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Tags: LongLeCE/llama.cpp

Tags

b9913

Toggle b9913's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Add Q2_0 quantization: type definition and CPU backend (ggml-org#24448)

b9892

Toggle b9892's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
common: Set optimal default thread count for ppc ( linux as well as A…

…IX) (ggml-org#25237)

b9878

Toggle b9878's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Fix stale tensor-split params for draft models (ggml-org#24814)

* meta: fix tensor split metadata for GQA attention

* Tidied the code a bit to match existing style

* Revert "Tidied the code a bit to match existing style"

This reverts commit b90c6c6.

* Reverted the ggml-backend-meta asset hack.

b9873

Toggle b9873's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
llama : add guard for K/V rotation input when buffer is unallocated (g…

…gml-org#25215)

llm_graph_input_attn_kv::set_input and llm_graph_input_attn_kv_iswa::set_input
call set_input_k_rot / set_input_v_rot whenever the rotation tensor pointer is
non-null, but the tensor's buffer can be unallocated (NULL) when a graph only
stores K/V without attending -- e.g. DFlash speculative decoding's KV-injection
pass. set_input_k_rot then calls ggml_backend_buffer_is_host() on a NULL buffer
and aborts with GGML_ASSERT(buffer).

Guard the four k_rot/v_rot inputs with the same "&& ->buffer" check that the
adjacent kq_mask inputs already use in these two functions. When the buffer is
unallocated there is no data to upload, so skipping is correct.

Fixes ggml-org#25191

Signed-off-by: liminfei-amd <[email protected]>

b9870

Toggle b9870's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
chat: trim messages sent to StepFun parser (fixes long reasoning loop…

…s) (ggml-org#25238)

* chat: trim messages sent to StepFun parser (fixes long reasoning loops)

* add regression test; remove duplicate template

* chat: trim StepFun content parts before rendering

The StepFun trim workaround ran on the already-rendered messages, where
typed content parts have been concatenated into a single string, so the
per-part whitespace could no longer be reached. Move the trim ahead of
rendering and apply it to content_parts text as well as the string
content and reasoning_content. Adds a content-parts regression test.

Co-Authored-By: Piotr Wilkin <[email protected]>
Assisted-By: Claude Fable 5 <[email protected]>

---------

Co-authored-by: tarruda <[email protected]>

b9860

Toggle b9860's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
llama : add llama_model_ftype_name() (ggml-org#25134)

* llama : add llama_model_ftype_name()

Expose the model file type (quantization) name, e.g. "Q8_0" or
"Q4_K - Medium", through a new public C API. The returned pointer is
valid for the lifetime of the model and nullptr when the model is
invalid or the file type is unknown.

Signed-off-by: Adrien Gallouët <[email protected]>

* Export enum

Signed-off-by: Adrien Gallouët <[email protected]>

* s/llama_model_ftype_name/llama_ftype_name/

Signed-off-by: Adrien Gallouët <[email protected]>

* Move "(guessed)" to the front in llama_ftype_name

Prepend the "(guessed)" label instead of appending it. This allows removing
the non-thread-safe static std::string, making the function allocation-free.

Signed-off-by: Adrien Gallouët <[email protected]>

* Add LLAMA_FTYPE_PREFIX

Signed-off-by: Adrien Gallouët <[email protected]>

* Dont check for model

Signed-off-by: Adrien Gallouët <[email protected]>

---------

Signed-off-by: Adrien Gallouët <[email protected]>

b9859

Toggle b9859's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
opencl: allow loading precompiled binary kernels from library (ggml-o…

…rg#23042)

* opencl: allow loading binary kernel

* opencl: add libdl.h

* ggml-backend-dl is in ggml, which depends backend libs, thus
  ggml-opencl cannot depend on ggml-backend-dl
* add libdl.h to break cyclic dep

* opencl: allow loading bin kernel lib

* opencl: load `gemm_moe_mxfp4_f32_ns` from kernel lib if available

* opencl: load q8_0 gemm from kernel lib

* opencl: load q4_0 moe gemm from kernel lib

* opencl: load q4_1 moe gemm from kernel lib

* opencl: load q4_k moe gemm from kernel lib

* opencl: always declare `get_adreno_bin_kernel_func_t`

* opencl: rephrase message

* opencl: fix for rebase

* opencl: update doc

b9851

Toggle b9851's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
cuda : prevent integer truncation and overflow errors when using KQ m…

…ask strides in flash_attn_mask_to_KV_max kernel (ggml-org#24945)

Co-authored-by: Stanisław Szymczyk <[email protected]>

b9843

Toggle b9843's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Revert "sched : reintroduce less synchronizations during split compute (

ggml-org#20793)" (ggml-org#25138)

b9838

Toggle b9838's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
common : remove unused regex-partial (ggml-org#25118)