ggml-org/llama.cpp
0
0 Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-10-21 09:09:40 +08:00
Code Issues Packages Projects Releases Wiki Activity
11,783 Commits 878 Branches 8,073 Tags

Default Branch

master

339f3242c7 · opencl: fix rope program build error with newer Clang (#30893) · Updated 2026-10-21 08:19:08 +08:00
 ZIP  TAR.GZ

Branches

gg/meta-forward-graph-optimize

8fc3cee2ac · meta : forward graph_optimize to underlying backends · Updated 2026-09-17 00:15:53 +08:00
712
1
 ZIP  TAR.GZ
qwen4exp-hc-fuse

26276bc99e · qwen4exp: add hc ops · Updated 2026-09-14 21:32:26 +08:00
145
1
 ZIP  TAR.GZ
alde/refactor-json-schema-qwen3

5746cc7b8c · cont : clean up grammar a bit · Updated 2026-09-13 05:14:54 +08:00
166
2
 ZIP  TAR.GZ
alde/refactor-json-schema

3a3f76f2e8 · cont : rename common_schema to common_chat_schema · Updated 2026-09-13 02:23:12 +08:00
169
27
 ZIP  TAR.GZ
xsn/download_flock

3c0c071cbf · simplify · Updated 2026-09-12 21:35:10 +08:00
171
2
 ZIP  TAR.GZ
gg/rpc-tensor

a70de036ff · rpc : add rdma_cm setup path for iWARP · Updated 2026-09-12 20:50:02 +08:00
170
3
 ZIP  TAR.GZ
xsn/server_refactor_router_subproc

8d79143a4d · Revert "tests: serialize the router tests that download the same model" · Updated 2026-09-12 05:31:38 +08:00
598
6
 ZIP  TAR.GZ
aman/qwen-sparse

41a4ad00dd · CUDA: enable sparse fa for qwen4 · Updated 2026-09-11 23:06:46 +08:00
282
1
 ZIP  TAR.GZ
aman/fix-offload-bug

3c6409920f · ggml: skip 0-sized ids tensor when offloading selected experts · Updated 2026-09-11 15:39:39 +08:00
282
1
 ZIP  TAR.GZ
xsn/mtmd_video_id

e5d72461ae · mtmd: propagate video ID to bitmap · Updated 2026-09-08 17:54:16 +08:00
657
1
 ZIP  TAR.GZ
cross-profiler

89f2417d62 · Apply batched suggestions from code review · Updated 2026-09-08 16:43:25 +08:00
263
2
 ZIP  TAR.GZ
xsn/log_jsonl_fit

7275941a06 · fit: add demo LOG_JSON · Updated 2026-09-08 05:46:16 +08:00
666
2
 ZIP  TAR.GZ
pr-24546-recreated

f01a75498e · CUDA: pick MMQ tile size against ncols_opt set on the host side · Updated 2026-09-07 22:39:17 +08:00
666
2
 ZIP  TAR.GZ
revert-mmq-moe-ncols

2265b7fa41 · Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" · Updated 2026-09-07 22:19:36 +08:00
252
1
 ZIP  TAR.GZ
xsn/models_fix_hang

eee2fa9c79 · server: keep a queued model out of the victim pool until its waiters leave · Updated 2026-09-07 19:39:02 +08:00
644
2
 ZIP  TAR.GZ
aman/draft-single-device

921c3e1aac · llama-context: add warning if sm tensor is used with one device · Updated 2026-09-06 20:21:21 +08:00
587
5
 ZIP  TAR.GZ
aman/cuda-fix-races

f0f9902d4c · use ggml_cuda_syncwarp · Updated 2026-09-06 16:08:55 +08:00
663
2
 ZIP  TAR.GZ
allozaur/ui/chat-perf

1cd349c00d · refactor : address review remarks · Updated 2026-09-06 09:29:20 +08:00
280
15
 ZIP  TAR.GZ
nikwen/m2-max-tunings

5150132b9a · metal : add remaining fa-vec tunings for M2 Max · Updated 2026-09-06 05:16:07 +08:00
280
1
 ZIP  TAR.GZ
xsn/log_jsonl

4c31a6aac2 · rename unknown to none · Updated 2026-09-06 03:52:33 +08:00
662
2
 ZIP  TAR.GZ