Skip to content
GitLab
Explore
Sign in
Tags
Tags give the ability to mark specific points in history as being important
b2481
76aa30a2
·
Add ability to use Q5_0, Q5_1, and IQ4_NL for quantized K cache (#6183)
·
Mar 21, 2024
b2480
c5b8595e
·
Add nvidia and amd backends (#6157)
·
Mar 21, 2024
b2479
42e21c68
·
cuda : fix conflict with std::swap (#6186)
·
Mar 21, 2024
b2478
1c51f98a
·
cuda : print the returned error when CUDA initialization fails (#6185)
·
Mar 20, 2024
b2476
272935b2
·
llava : add MobileVLM_V2 backup (#6175)
·
Mar 20, 2024
b2475
ccf58aa3
·
cuda : refactor to remove global resources (#6170)
·
Mar 20, 2024
b2474
91f8ad16
·
Server: version bump for httplib and json (#6169)
·
Mar 20, 2024
b2471
d795988d
·
Revert "llava : add a MobileVLM_V2-1.7B backup (#6152)"
·
Mar 20, 2024
b2466
d26e8b66
·
increase igpu cluster limit (#6159)
·
Mar 20, 2024
b2465
d8b009a9
·
Remove undeed header file. (#6158)
·
Mar 19, 2024
b2463
b80cf3b2
·
common : disable repeat penalties by default (#6127)
·
Mar 19, 2024
b2462
970a4806
·
ci : exempt some labels from being tagged as stale (#6140)
·
Mar 19, 2024
b2461
4c28b825
·
common : print usage on '-h' and '--help' (#6145)
·
Mar 19, 2024
b2459
d199ca79
·
mpt : implement backwards compatiblity with duped output tensor (#6139)
·
Mar 18, 2024
b2458
104f5e0f
·
clip : fix memory leak (#6138)
·
Mar 18, 2024
b2457
5e1b7f94
·
backend : set max split inputs to GGML_MAX_SRC (#6137)
·
Mar 18, 2024
b2456
ac9ee6a4
·
ci : disable stale issue messages (#6126)
·
Mar 18, 2024
b2455
4f6d1337
·
ci : temporary disable sanitizer builds (#6128)
·
Mar 18, 2024
b2454
2bf8d0f7
·
backend : offload large batches to GPU (#6083)
·
Mar 18, 2024
b2453
496bc79b
·
common : tidy-up argument parsing (#6105)
·
Mar 18, 2024
Prev
1
…
14
15
16
17
18
19
20
21
22
…
99
Next