ggerganov/llama.cpp

Pull Requests Commits
chore: clean useless beam search param (#7985)
thxCode committed 1 year ago
Verified b96f9afb
readme : update UI list (#7943)
abgulati committed 1 year ago
Verified 11937781
ggml : sync
ggerganov committed 1 year ago
Verified 5326bcce
whisper : use ggml_backend_sched (whisper/2239)
ggerganov committed 1 year ago
Verified e6ecc2be
update: support Qwen2-57B-A14B (#7835)
legraphista committed 1 year ago
Verified a94e6ff8
Make updates to type cast based on compiler instead of OS (#7851)
Srihari-mcw committed 1 year ago
Verified 5b6da187
llama : disable FA if KV head size do not match (#7982)
ggerganov committed 1 year ago
Verified 7c26775a
Add Nix and Flox install instructions (#7899)
bryanhonof committed 1 year ago
Verified b473e950
sched : offload_op also requires supports_op (#7977)
slaren committed 1 year ago
Verified 99052cd2
fix: divide 0 exception in mamba (#7932)
thxCode committed 1 year ago
Verified c637fcd3
Implement non-mapped async IO for CUDA on Windows. (#7896)
mtavenrath committed 1 year ago
Verified 6a2f0b34
rpc : fix load/store misaligned addresses (#7948)
ggerganov committed 1 year ago
Verified 21be9cab
gguf-dump.py: add --markdown dump output (#7853)
mofosyne committed 1 year ago
Verified 006167aa
[SYCL] Update README-sycl.md for Chapter "Recommended release" and "News" (#7946)
arthw committed 1 year ago
Verified df68d4fa
Add support for sqrt on CUDA (#7953)
calvin-laurenson committed 1 year ago
Verified 43b35e38
cuda : fix bounds check for src0 rows in MMVQ kernel (whisper/2231)
ggerganov committed 1 year ago
Verified 19b7a836
ggml : fix and optimize ppc64le (ggml/849)
penghongbo committed 1 year ago
Verified b5fcf8ef
ggml : remove duplicate include of ggml-common.h (ggml/853)
danbev committed 1 year ago
Verified 398105ff
flake.lock: Update (#7951)
ggerganov committed 1 year ago
Verified bc6c457f
unicode : avoid char32_t (#7957)
ggerganov committed 1 year ago
Verified 52399254
readme : update UI list [no ci] (#7958)
hopkins385 committed 1 year ago
Verified 6fe1c627
ggml : fix handling of zero blocks in IQ quants (#7955)
ggerganov committed 1 year ago
Verified cddaf028
github : update pr template
ggerganov committed 1 year ago
Verified c8a82194
Vulkan Shader Refactor, Memory Debugging Option (#7947)
0cc4m committed 1 year ago
Verified 7c7836d9
Add `cvector-generator` example (#7514)
ngxson committed 1 year ago
Verified 0c7b3595
[SYCL] remove global variables (#7710)
airMeng committed 1 year ago
Verified 7b2f4a7d
ci : fix macos x86 build (#7940)
olexiyb committed 1 year ago
Verified f8ec8877
CUDA: faster q2_K, q3_K MMQ + int8 tensor cores (#7921)
JohannesGaessler committed 1 year ago
Verified 76d66ee0
metal : utilize max shared memory for mul_mat_id (#7935)
ggerganov committed 1 year ago
Verified 66ef1cee
llama-bench : fix RPC indication (#7936)
rgerganov committed 1 year ago
Verified e65bbf60
Terms Of Use Privacy Policy Refund Policy Impressum