ngxson/llama.cpp

Pull Requests Commits
llava : Allow locally downloaded models for QwenVL (#10833)
bartowski1182 committed 1 year ago
Verified 4ddd199f
llama : add Deepseek MoE v1 & GigaChat models (#10827)
Inf1delis committed 1 year ago
Verified a0974156
scripts : change build path to "build-bench" for compare-commits.sh (#10836)
ggerganov committed 1 year ago
Verified 87cf323c
server: (UI) add syntax highlighting and latex math rendering (#10808)
VJHack committed 1 year ago
Verified 5478bbcd
gguf-py : bump to v0.13.0
ggerganov committed 1 year ago
Verified b5ae1ddf
server: Fix `has_next_line` in JSON response (#10818)
MichelleTanPY committed 1 year ago
Verified 89d604f2
nix: allow to override rocm gpu targets (#10794)
kurnevsky committed 1 year ago
Verified e52aba53
llama : add Qwen2VL support + multimodal RoPE (#10361)
HimariO committed 1 year ago
Verified ba1cb19c
Removes spurious \r in output that causes logging in journalctl to treat lines as binary and therefore hidden by default (#10771)
cduk committed 1 year ago
Verified 56eea078
Introducing experimental OpenCL backend with support for Qualcomm Adreno GPUs (#10693)
lhez committed 1 year ago
Verified a76c56fa
Opt class for positional argument handling (#10508)
ericcurtin committed 1 year ago
Verified c27ac678
fix: graceful shutdown for Docker images (#10815)
co42 committed 1 year ago
Verified 11e07fd6
gguf-py : numpy 2 newbyteorder fix (#9772)
jettjaniak committed 1 year ago
Verified 4601a8bb
update docs
ngxson committed 1 year ago
630ddcc5
Merge branch 'master' into xsn/fix_logprobs
ngxson committed 1 year ago
262950db
add --multi-token-probs
ngxson committed 1 year ago
196e237e
Fix crash caused by ggml_backend_load_all when launching on Android Activity (#10812)
sienaiwun committed 1 year ago
Verified 9f35e445
vulkan: small mul_mat_vec optimizations (#10665)
netrunnereve committed 1 year ago
Verified 64ae0655
update docs
ngxson committed 1 year ago
06bb38e7
SYCL: Reduce most of the compiler warnings (#10748)
qnixsynapse committed 1 year ago
Verified 83ed24a9
ggml : Fix compilation issues on ARM platform when building without fp16 (#10811)
kkontny committed 1 year ago
Verified d583cd03
common : improve -ctv -ctk CLI arguments (#10806)
ngxson committed 1 year ago
Verified adffa6ff
contrib : add ngxson as codeowner (#10804)
ngxson committed 1 year ago
Verified 274ec65a
CUDA: faster non-contiguous concat (#10760)
A3shTnT committed 1 year ago
Verified 8faa1d4d
remove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS (#10797)
slaren committed 1 year ago
Verified cb13ef85
Vulkan: Use improved q4_k and q5_k dequant code in dequant shaders (#10798)
0cc4m committed 1 year ago
Verified 4064c0e3
Vulkan: Add VK_EXT_subgroup_size_control support to ensure full subgroups for coopmats (#10721)
0cc4m committed 1 year ago
Verified dc5301d5
common : add missing env var for speculative (#10801)
ngxson committed 1 year ago
Verified 9fdb1243
set p for sampled token
ngxson committed 1 year ago
ed7f2d57
fix test
ngxson committed 1 year ago
22b72c85
Terms Of Use Privacy Policy Refund Policy Impressum