Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

split out rllm common from rllm-cuda #51

Merged
merged 56 commits into from
Feb 12, 2024
Merged

split out rllm common from rllm-cuda #51

merged 56 commits into from
Feb 12, 2024

Conversation

mmoskal
Copy link
Member

@mmoskal mmoskal commented Feb 6, 2024

WIP

This converts from #ifdef style to traits

Fixes #48

mmoskal added 30 commits February 5, 2024 21:39
@mmoskal
start #ifdef (cfg) -> trait conversion
d17dfc2
@mmoskal
fix warnings
119af1c
@mmoskal
force GPU layers when --cuda on llama.cpp
4b74440
@mmoskal
parametric RllmConfig
3f39e75
@mmoskal
splitting out model loader args
0d388f5
@mmoskal
add sequence manager
19210e0
@mmoskal
cleanup
547b161
@mmoskal
use traits in tch version
83f55fd
@mmoskal
fix warnings
662fd23
@mmoskal
move paged into llm/
b0254f8
@mmoskal
limit usage of llm::
e8e32d1
@mmoskal
cleanup
594c8b0
@mmoskal
default unique --shm-prefix
9d336b3
@mmoskal
keep block mapping in BlockSpaceManager
535f5a0
@mmoskal
make cpp version build again
832abbd
@mmoskal
feature-containment
f2e22d7
@mmoskal
remove last #[cfg]
12ae462
@mmoskal
split cpp/cuda code into folders
469cf64
@mmoskal
make it compile again
6c50da4
@mmoskal
fix warnings
385e8c3
@mmoskal
fix start script
232d720
@mmoskal
disable ccache action (runs out of space)
e0c4d39
@mmoskal
rename rllm-server to rllm-cuda
6a233b3
@mmoskal
custom md link check
0ae83cc
@mmoskal
Merge remote-tracking branch 'origin/main' into traits
d390f1c
@mmoskal
fix links
07b5e12
@mmoskal
fix build script
ef7c432
@mmoskal
disable all build caching
90fcc46
@mmoskal
diagnose space problem
bd915b6
@mmoskal
update llama.cpp
7de3e0b
mmoskal added 26 commits February 7, 2024 16:04
@mmoskal
try a hack for M1
963dffa
@mmoskal
disable failing cuda build
0cc32ed
@mmoskal
Revert "disable all build caching"
b0bf6ba 
This reverts commit 90fcc46.
@mmoskal
remove unused deps
8bf1ddb
@mmoskal
start moving files around; fixes #56
3434490
@mmoskal
update some file refs
1597110
@mmoskal
more folder renames
729a223
@mmoskal
updating links
16adf25
@mmoskal
more path updates
5ef6435
@mmoskal
cpp-server.sh -> server.sh
66de406
@mmoskal
Fix typo in REST.md
3e4fd97
@mmoskal
fix path
e46c772
@mmoskal
fix folders in release script
5ced06d
@mmoskal
fix #46: GPU acceleration on macOS
b30dd81
@mmoskal
switch to upstream rquickjs
73672c7 
(merged DelSkayn/rquickjs#263 )
@mmoskal
tone down CUDA instructions in the guide
4ab588a
@mmoskal
explain --cuda vs rllm-cuda
cc5c20a
@mmoskal
Add check for Cargo.toml existence in build_rust function
2fc7e40
@mmoskal
add tps stats
84a099e
@mmoskal
allow reordering server.sh args
3e5fce3
@mmoskal
upload .wasm files directly in release
7a056e3
@mmoskal
start on gh controllers; fixes #55
c8f214c
@mmoskal
implement module download
4326b45
@mmoskal
better error handling
ff1d3b5
@mmoskal
fix rllm-cuda phi2 crash
76ac24b
@mmoskal
have aici cli use gh: controllers
5e944b6
@mmoskal mmoskal merged commit 5e944b6 into main Feb 12, 2024
7 checks passed
@mmoskal mmoskal deleted the traits branch February 12, 2024 18:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

split out common rllm from rllm-cuda
1 participant
@mmoskal