Daniel Hiltgen
|
e1f50377f4
Harden generate patching model
|
1 year ago |
Daniel Hiltgen
|
e02ecfb6c8
Merge pull request #2116 from dhiltgen/cc_50_80
|
1 year ago |
Jeffrey Morgan
|
a64570dcae
Fix clearing kv cache between requests with the same prompt (#2186)
|
1 year ago |
Daniel Hiltgen
|
a447a083f2
Add compute capability 5.0, 7.5, and 8.0
|
1 year ago |
Jeffrey Morgan
|
4c54f0ddeb
sign dylibs on macOS (#2101)
|
1 year ago |
Jeffrey Morgan
|
dc88cc3981
use `gzip` for runner embedding (#2067)
|
1 year ago |
Daniel Hiltgen
|
1b249748ab
Add multiple CPU variants for Intel Mac
|
1 year ago |
Jeffrey Morgan
|
288ef8ff95
add `gcc -lstdc++` flag for linux cpu (#1974)
|
1 year ago |
Jeffrey Morgan
|
4cf17990f7
use g++ to build `libext_server.so` on linux (#1972)
|
1 year ago |
Daniel Hiltgen
|
d88c527be3
Build multiple CPU variants and pick the best
|
1 year ago |
Bruce MacDonald
|
3367b5f3df
remove unused generate patches (#1810)
|
1 year ago |
Daniel Hiltgen
|
9983fa5f4e
Cleaup stale submodule
|
1 year ago |
Daniel Hiltgen
|
77d96da94b
Code shuffle to clean up the llm dir
|
1 year ago |