slaren
Follow
Popular repositories Loading
-
llama.cpp
llama.cpp PublicForked from ggml-org/llama.cpp
Port of Facebook's LLaMA model in C/C++
C++ 10
-
-
-
-
-
Hearthstone-Deck-Tracker
Hearthstone-Deck-Tracker PublicForked from HearthSim/Hearthstone-Deck-Tracker
C#
1,516 contributions in the last year
Less
No contributions.
Low contributions.
Medium-low contributions.
Medium-high contributions.
High contributions.
More
Contribution activity
March 2025
Created 5 commits in 3 repositories
Created a pull request in ggml-org/llama.cpp that received 5 comments
ggml-backend : keep paths in native string type when possible
Avoids conversions from std::filesystem::path to non-native std::string to prevent possible issues when converting between different encodings. Whe…
+50
−54
lines changed
•
5
comments
Opened 1 other pull request in 1 repository
ggml-org/llama.cpp
1
merged
-
test-backend-ops : add option -p to filter by op params
This contribution was made on Mar 2
Reviewed 26 pull requests in 3 repositories
ggml-org/llama.cpp
23 pull requests
-
Simplify and improve CUDA graphs through use of indirect copy pointers
This contribution was made on Mar 25
-
ggml : fix quantized cpy op
This contribution was made on Mar 21
-
model : do not repack if a GPU device is present
This contribution was made on Mar 21
-
Add PLM GGUF Conversion & Inference Support
This contribution was made on Mar 19
-
Fix backend search path
This contribution was made on Mar 11
-
CUDA: compress-mode size
This contribution was made on Mar 11
-
llama : refactor llama_context, llama_kv_cache, llm_build_context (v2)
This contribution was made on Mar 11
-
clip : bring back GPU support
This contribution was made on Mar 10
-
musa: support new arch mp_31 and update doc
This contribution was made on Mar 10
-
tests : fix test-quantize-fns to init the CPU backend
This contribution was made on Mar 10
-
tests: use adaptive number of threads
This contribution was made on Mar 8
-
ggml-backend : make path_str compatible with C++20
This contribution was made on Mar 8
-
CUDA: fix FA logic for PTX 7.0 and CC >= 7.5
This contribution was made on Mar 6
-
ggml-cpu: faster AVX2 variant for IQ1_M
This contribution was made on Mar 6
-
HIP: rocWMMA documentation and enabling in workflow builds
This contribution was made on Mar 6
-
HIP/CUDA: set the paramerter value in maintain_cuda_graph instead of replaceing it.
This contribution was made on Mar 6
-
opencl: Fix not enough space in the buffer
This contribution was made on Mar 6
-
opencl: Fix
ulongkernel args were set fromintvariablesThis contribution was made on Mar 6 -
opencl:Fix profile-related errors
This contribution was made on Mar 6
-
ggml-cpu: Faster IQ1 mul_mat_vec on AVX2 using BMI2 instructions
This contribution was made on Mar 6
-
ci : add fetch-depth to xcframework upload
This contribution was made on Mar 5
-
Some portability improvements from trying to build with Visual Studio 2017
This contribution was made on Mar 3
-
Fix kleidiai build
This contribution was made on Mar 3
ggml-org/ggml
2 pull requests
-
cmake: Enable specifying exact PowerPC CPU architecture
This contribution was made on Mar 10
-
cuda: unary ops - perform the final computation as a float, and de-duplicate the code
This contribution was made on Mar 3
ggerganov/whisper.cpp
1 pull request
-
ci : remove CMAKE_CUDA_ARCHITECTURES in windows-cublas
This contribution was made on Mar 22
2
contributions
in private repositories
Mar 11