ngxson/llama.cpp

Pull Requests Commits
model-conversion : remove -c 0 from model card template [no ci] (#18807)
danbev committed 36 days ago
Verified 20ca2e12
HIP: add fattn-mma-f16 for RDNA4 (#18481)
zhang-hui-yulo committed 36 days ago
Verified ea4a321f
doc: ban AI-generated PR descriptions [no ci] (#18765)
JohannesGaessler committed 36 days ago
Verified c1e79e61
mtmd: fix use_non_causal being reported incorrectly (#18793)
ngxson committed 36 days ago
Verified e047f9ee
CUDA : fix unused argument when USE_CUDA_GRAPH=OFF (#18800)
ggerganov committed 36 days ago
Verified 0a57271a
graph : clean up t5 input builders (#18795)
gabe-l-hart committed 36 days ago
Verified 076b0faf
llama-bench: add direct_io parameter (#18778)
0cc4m committed 36 days ago
Verified db79dc06
clean up
ngxson committed 37 days ago
9481b9d2
ci : remove libcurl in releases (#18775)
angt committed 37 days ago
Verified 537d4240
fix
ngxson committed 37 days ago
25702ba8
server : add arg for disabling prompt caching (#18776)
rgerganov committed 37 days ago
Verified bcf75461
add comment pointing to this PR
ngxson committed 37 days ago
821e329e
fix
ngxson committed 37 days ago
e32545f1
server : make sure children tasks are scheduled to launch with parent
ngxson committed 37 days ago
b55964a1
ci : use openssl for openEuler-latest-cmake-cann (#18779)
angt committed 37 days ago
Verified 36c5913c
vendor : update cpp-httplib to 0.30.1 (#18771)
angt committed 37 days ago
Verified 8e649571
examples : add --kv-unified to batched example (#18774)
danbev committed 37 days ago
Verified 4150da9a
vulkan: change memory_logger to be controlled by an env var (#18769)
jeffbolznv committed 37 days ago
Verified 8e2da778
server: update docs for sleeping [no ci] (#18777)
ngxson committed 37 days ago
Verified ce3bf9b1
vulkan: Use VK_EXT_shader_64bit_indexing to handle large mat_mul(_id) (#18678)
jeffbolznv committed 37 days ago
Verified 2bbe4c2c
vulkan: Disable large coopmat matmul configuration on proprietary AMD driver (#18763)
0cc4m committed 37 days ago
Verified 1051ecd2
model: fix qwen3next broken due to #18683 (#18762)
ngxson committed 38 days ago
Verified 0c3b7a9e
Vulkan: Optimize Matmul parameters for AMD GPUs with Coopmat support (#18749)
0cc4m committed 38 days ago
Verified 0e76501e
security: make it clear about subtopics in server (#18754)
ngxson committed 38 days ago
Verified 4b060bf2
debug : include LLAMA_POOLING_TYPE_UNSPECIFIED in pooling check (#18692)
danbev committed 38 days ago
Verified 9789e284
tests : refactor test-backend-sampler (#18753)
ggerganov committed 38 days ago
Verified 84ae04f1
model: try to improve Qwen3 Next (#18683)
ngxson committed 38 days ago
Verified 506bb6e0
readme : update UIs (#18751)
thom-dev-fr committed 38 days ago
Verified 79456a69
security: narrow down the scope of what we consider a vulnerability (#18752)
ngxson committed 38 days ago
Verified 28068af7
opencl: add SOFTPLUS op support (#18726)
shaofeiqi committed 38 days ago
Verified 707cbafc
Terms Of Use Privacy Policy Refund Policy Impressum