4eb3a95cbb
reenable LCM sampler
2024-03-05 17:39:21 +08:00
ac43e0115c
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# .devops/nix/package.nix
# README.md
# ggml-metal.m
# llama.cpp
# scripts/sync-ggml.last
# tests/test-backend-ops.cpp
2024-03-05 15:54:05 +08:00
c952b4f192
Revert "merge missing functions from sdcpp"
...
This reverts commit 19e1c518f1
2024-03-05 15:38:51 +08:00
0c59c1ed90
allow specifying width and height
2024-03-03 15:44:15 +08:00
e53d21d748
sanitize SD prompt to avoid segfault
2024-03-02 12:05:59 +08:00
2d9a90b652
try to fix ci compile errors (+1 squashed commits)
...
Squashed commits:
[d0d49663] fixed log multiline (+1 squashed commits)
Squashed commits:
[81a8befe] try to fix linux build error (+1 squashed commits)
Squashed commits:
[22850dda] try to fix build (+1 squashed commits)
Squashed commits:
[b8294611] missing type
2024-03-01 23:38:15 +08:00
80011ed8aa
KCPP SD: add warn and step restriction., updated lite, handle quant mode
2024-03-01 16:41:19 +08:00
3463688a0e
image generation is fully working over api (+1 squashed commits)
...
Squashed commits:
[c98ab0b4] single image generation is working now
2024-03-01 14:43:44 +08:00
e8f4d7b3da
added model and config endpoints for sdcpp, added more samplers. speed is still not good
2024-02-29 22:56:09 +08:00
257015bb94
Resolve Metal compilation errors for sdcpp ( #720 )
2024-02-29 20:15:45 +08:00
5a44d4de2b
refactor and clean identifiers for sd, fix cmake
2024-02-29 18:28:45 +08:00
66134bb36e
ui for loading SD models done
2024-02-29 17:08:22 +08:00
524ba12abd
refactor - do not use a copy buffer to store generation outputs, instead return a cpp allocated ptr
2024-02-29 14:02:20 +08:00
f75e479db0
WIP on sdcpp integration
2024-02-29 00:40:07 +08:00
8a919daafb
add as library to makefile
2024-02-28 17:45:00 +08:00
17355faf6e
sdcpp is working!
2024-02-28 17:00:18 +08:00
19e1c518f1
merge missing functions from sdcpp
2024-02-28 16:38:00 +08:00
26696970ce
initial files from sdcpp (not working)
2024-02-28 15:45:13 +08:00
ad638285de
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# Makefile
# README.md
# flake.lock
# ggml-cuda.cu
# llama.cpp
# tests/test-backend-ops.cpp
# tests/test-quantize-fns.cpp
2024-02-28 13:41:35 +08:00
7eccc5ffa6
change listen count, fix null
2024-02-16 16:01:24 +08:00
f9bc7245ab
b64 decoder
2024-02-11 20:35:34 +08:00
acb792815e
try fix cuda slowdown
2024-02-05 16:34:15 +08:00
35c32fd0f2
refactor some old code with batching
2024-02-05 15:54:45 +08:00
5639c1a520
units (+2 squashed commit)
...
Squashed commit:
[166979d9] units coversion
[038dd5d4] get rid of all warnings (+1 squashed commits)
Squashed commits:
[6efd1e1b] get rid of all warnings
2024-01-20 23:53:21 +08:00
71e9a64171
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# .github/workflows/nix-ci.yml
# CMakeLists.txt
# Makefile
# ggml-cuda.cu
# ggml-opencl.cpp
# llama.cpp
2024-01-20 23:27:42 +08:00
a137b6b9ff
fixed typo
2024-01-20 17:28:14 +08:00
680a41ed71
refactor identifiers
2024-01-20 17:26:11 +08:00
693f3f0b00
try to use allocator for cuda ggml v3
2024-01-20 12:53:31 +08:00
97693e7e97
increase pool buffers
2024-01-20 11:52:39 +08:00
21f0ce2502
do not use vmm for ggml v3
2024-01-20 11:30:22 +08:00
db14de5c32
fossilize ggml library ver 3, to support ggjtv3
2024-01-20 10:49:25 +08:00
230a638512
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# .github/workflows/docker.yml
# CMakeLists.txt
# Makefile
# README.md
# llama.cpp
# tests/test-grad0.cpp
2023-12-22 14:40:13 +08:00
76a3ba42eb
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# ggml.c
# ggml.h
# requirements.txt
# tests/test-quantize-perf.cpp
2023-12-16 22:58:53 +08:00
a6eb9b8010
Fix GPT2 not loading due to graph too small
2023-11-26 23:06:42 +08:00
dc4078c039
fixed segfault with all non-gguf models
2023-11-20 22:31:56 +08:00
45ad1b97f8
max nodes 8192
2023-11-18 11:02:35 +08:00
35a97e14b2
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# Makefile
# README.md
# docs/token_generation_performance_tips.md
# grammars/README.md
# scripts/sync-ggml.sh
# tests/CMakeLists.txt
# tests/test-grad0.cpp
# tests/test-opt.cpp
2023-11-15 16:59:53 +08:00
8c14c81b33
hopefully this fixes the dotnet nonsense
2023-11-03 11:23:56 +08:00
42eabf2f2f
rope fixes
2023-11-02 20:41:16 +08:00
bc4ff72317
not working merge
2023-11-02 17:52:40 +08:00
cc5b282350
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# CMakeLists.txt
# Makefile
# build.zig
# flake.lock
# flake.nix
# ggml.c
2023-10-31 20:44:04 +08:00
5db89b90b7
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# .gitignore
# CMakeLists.txt
# Makefile
# README.md
# build.zig
# ggml-opencl.cpp
# tests/CMakeLists.txt
# tests/test-double-float.cpp
# tests/test-sampling.cpp
2023-10-25 23:58:15 +08:00
c9983a72d6
prevent lora with clblast
2023-10-25 15:18:03 +08:00
f288c6b5e3
Merge branch 'master' into concedo_experimental
...
# Conflicts:
# CMakeLists.txt
# Makefile
# build.zig
# scripts/sync-ggml.sh
2023-10-10 00:09:46 +08:00
5b4cef5a60
archived old unused file
2023-10-02 16:57:20 +08:00
202e28a76a
do not offload rope for old cublas (+1 squashed commits)
...
Squashed commits:
[ca72a66f] fix allocr (+1 squashed commits)
Squashed commits:
[22a0e30e] updated lite
2023-09-30 18:18:36 +08:00
b84e210f0d
merge new rope param nonsense
2023-09-30 11:33:30 +08:00
f6ba36dff6
Reduce warnings. ( #439 )
2023-09-16 18:52:09 +08:00
5925c23d51
fix for RWKV
2023-09-01 17:02:11 +08:00
d7fed4732f
fix for typical sampler
2023-09-01 15:24:00 +08:00
Concedo
bebopkim
Ycros