ggml-org/llama.cpp

Pull Requests Commits
kv-cache: honor PARTIAL_ONLY in the plain KV cache (#30514)
ServeurpersoCom committed 6 days ago
Verified e467e5b5
graph: keep the graph non empty without embeddings (#30522)
ServeurpersoCom committed 6 days ago
Verified 1ddc2625
server: save only restorable slot states to the prompt cache (#30498)
ServeurpersoCom committed 6 days ago
Verified 2639a3b4
spec : add lilicorr support (#30506)
frozenblade1224 committed 6 days ago
Verified 843e7dcc
vulkan: Use L2-friendly workgroup scheduling for cm2 mulmat (#30215)
0cc4m committed 6 days ago
Verified 3e390af1
opencl: support CLAMP on non-contiguous views (#30186)
wanghqc committed 6 days ago
Verified a0e6dc73
devops: incorrect COPY instruction (#30529)
stormyhs committed 7 days ago
Verified 23deec02
gguf-py : detect nfc normalization (#30540)
CISC committed 7 days ago
Verified 683bbc0d
sycl: IQ3 code reorder (#29107)
anantshri committed 7 days ago
Verified 529ac893
ggml-cpu : keep the FP16->FP32 lookup table on x86 with F16C (#30342) (#30343)
Tough-Respawn committed 7 days ago
Verified 685f7089
docs: update IBM zDNN backend docs (#30527)
taronaeo committed 7 days ago
Verified c114bd4c
ggml-cpu : size set_rows conversion buffer for all threads (#30454)
am17an committed 7 days ago
Verified 5cdadfc3
cuda, hip: GDN chunked kernel (#29353)
am17an committed 7 days ago
Verified 62df1815
skill: Update `code-review` skill for UI development (#30509)
allozaur committed 7 days ago
Verified 37d6d653
ggml: fix conv_1d result layout for batched input (#28738)
ServeurpersoCom committed 7 days ago
Verified 6868a52c
common : add --spec-draft-n-ubatch (#30459)
ggerganov committed 7 days ago
Verified 3fb47677
docs : clarify installation recommendation for Intel oneAPI Toolkit (#30407)
alosslessdev committed 7 days ago
Verified 318a471a
ggml : add note about mul_mat_id + disable mul_mat_id_dup tests (#30501)
ggerganov committed 7 days ago
Verified 8c2134a8
model: split layers evenly when no device reports free memory (#29677)
ServeurpersoCom committed 7 days ago
Verified df18bee1
vendor : update cpp-httplib to 0.61.0 (#30494)
angt committed 7 days ago
Verified 3eb10136
ggml : improve RPC tensor transfer logs (#30481)
ggerganov committed 7 days ago
Verified af0579c7
cuda : use SIMT implicit GEMM conv2d on AMD without WMMA (#30325)
aisk committed 7 days ago
Verified 62bacfa0
jinja : write floats in tojson as Python does (#30362)
ykhrustalev committed 7 days ago
Verified 77390baa
opencl: use buf+offset instead of subbuffer for adreno moe expert sort (#30418)
lhez committed 7 days ago
Verified c5a99762
opencl: fix q8_0 dp4a mul_mat_id and mark PREC_F32 mul_mat_id unsupported for adreno (#30365)
lhez committed 7 days ago
Verified 21a19cd3
models: drop the t_embd expand of the embedding graphs (#30461)
ServeurpersoCom committed 7 days ago
Verified 9cf967a0
HIP: use 64-row MMQ tiles for some quant types on RDNA3.5 (#28195)
liangliangchang committed 7 days ago
Verified 4dd6dd4b
vulkan: fix compile error (#30495)
0cc4m committed 7 days ago
Verified 14310190
llama : build the k-pool selection in the causal mask layout (#30480)
ggerganov committed 7 days ago
Verified e7268bf6
vulkan: Add workaround for device free memory > total memory on Intel Windows (#29835)
rillomas committed 7 days ago
Verified cdce056a
Terms Of Use Privacy Policy Refund Policy Impressum