llama.cpp
llama : save and restore kv cache for single seq id
#6341
Merged

llama : save and restore kv cache for single seq id #6341

kaetemi
kaetemi llama : save and restore kv cache for single seq id
662aaea8
kaetemi remove trailing whitespace
54628178
kaetemi respond error in case there's no space in the kv cache
ab1c46a7
kaetemi add kv seq save restore to test case
02a18406
martindevans
compilade compilade requested a review from compilade compilade 1 year ago
kaetemi add --slot-save-path arg to enable save restore and restrict save loc…
b8e8facb
kaetemi
martindevans
kaetemi
martindevans
kaetemi
martindevans
martindevans Returning 0 for some cases, instead of asserting.
b182f8f6
martindevans
kaetemi cleanup error cases
a2b48b95
ggerganov
ggerganov commented on 2024-03-28
kaetemi rename sequence state functions
c4443d7a
kaetemi rename state get set functions
4d5356bb
phymbert
phymbert requested changes on 2024-03-29
kaetemi add previous function names back in with DEPRECATED notice
bbcbf47b
kaetemi update doc
8b5ae299
kaetemi adjust endpoints to preferred style
a71ec3db
kaetemi fix restoring zero cell count
bf1d4932
kaetemi handle seq rm return value
8ab1a172
kaetemi unused param
0d221367
kaetemi keep in the size check
29f18c29
kaetemi fix return types
f2e41b32
kaetemi add server test case for slot save restore
92c46810
kaetemi cleanup
60f685ff
compilade
compilade commented on 2024-03-29
phymbert
phymbert commented on 2024-03-30
phymbert
phymbert requested changes on 2024-03-30
kaetemi add cake
d38eef46
kaetemi
kaetemi cleanup style
ea717f77
kaetemi add special
b509b8b3
kaetemi removing a whole sequence never fails
129b6ffe
kaetemi move sequence state file functionality from server to llama to match …
8af72118
phymbert
phymbert approved these changes on 2024-03-31
slaren
slaren commented on 2024-03-31
kaetemi catch exceptions on save as well
3d6fa5bd
kaetemi error log messages
b3f6da3d
kaetemi check types for stricter restore
be714a0f
kaetemi update server doc
0ccfbf2f
kaetemi kaetemi requested a review from ggerganov ggerganov 1 year ago
ggerganov
ggerganov approved these changes on 2024-04-04
ggerganov readme : update API changes date
205c44c2
ggerganov
kaetemi Merge branch 'master' into feature/save-restore-seq
d9fd0d7e
kaetemi
github-actions
ngxson
ngxson commented on 2024-04-04
ggerganov ggerganov requested a review from ngxson ngxson 1 year ago
kaetemi strict filename validation
f2a4777d
kaetemi move include, reject bom as well
4a4f3993
kaetemi also reject empty filename
2fbf0c34
kaetemi reject whitespace and trailing dot
bf94e9f7
ngxson
ngxson approved these changes on 2024-04-06
kaetemi kaetemi requested a review from ggerganov ggerganov 1 year ago
ggerganov ggerganov merged beea6e1b into master 1 year ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone