Bruce MacDonald
|
326de48930
use loaded llm for embeddings
|
1 éve |
Patrick Devine
|
d9cf18e28d
add maximum retries when pushing (#334)
|
1 éve |
Michael Yang
|
6517bcc53c
Merge pull request #290 from jmorganca/add-adapter-layers
|
1 éve |
Michael Yang
|
6a6828bddf
Merge pull request #167 from jmorganca/decode-ggml
|
1 éve |
Jeffrey Morgan
|
040a5b9750
clean up cli flags
|
1 éve |
Michael Yang
|
6de5d032e1
implement loading ggml lora adapters through the modelfile
|
1 éve |
Michael Yang
|
fccf8d179f
partial decode ggml bin for more info
|
1 éve |
Bruce MacDonald
|
4b3507f036
embeddings endpoint
|
1 éve |
Bruce MacDonald
|
868e3b31c7
allow for concurrent pulls of the same files
|
1 éve |
Bruce MacDonald
|
09d8bf6730
fix build errors
|
1 éve |
Bruce MacDonald
|
7a5f3616fd
embed text document in modelfile
|
1 éve |
Jeffrey Morgan
|
cff002b824
use content type `application/x-ndjson` for streaming responses
|
1 éve |
Jeffrey Morgan
|
a027a7dd65
add `0.0.0.0` as an allowed origin by default
|
1 éve |
Bruce MacDonald
|
21ddcaa1f1
pr comments
|
1 éve |
Michael Yang
|
f2074ed4c0
Merge pull request #306 from jmorganca/default-keep-system
|
1 éve |
Bruce MacDonald
|
a6f6d18f83
embed text document in modelfile
|
1 éve |
Michael Yang
|
4dc5b117dd
automatically set num_keep if num_keep < 0
|
1 éve |
cmiller01
|
fb593b7bfc
pass flags to `serve` to allow setting allowed-origins + host and port
|
1 éve |
Jeffrey Morgan
|
e3fb1fd3f1
server: compare options correctly
|
1 éve |
Bruce MacDonald
|
8b1e791820
allow specifying zero values in modelfile
|
1 éve |
Jeffrey Morgan
|
03cff3a225
server: reset digest at end of generate
|
1 éve |
Bruce MacDonald
|
8f8b6288ac
check server is running before running command
|
1 éve |
Bruce MacDonald
|
765994362c
use head to check heartbeat
|
1 éve |
Bruce MacDonald
|
1c5a8770ee
read runner parameter options from map
|
1 éve |
Bruce MacDonald
|
daa0d1de7a
allow specifying zero values in modelfile
|
1 éve |
Jeffrey Morgan
|
528bafa585
cache loaded model
|
1 éve |
Bruce MacDonald
|
671eec6da9
log prediction failures
|
1 éve |
Michael Yang
|
f62a882760
add session expiration
|
1 éve |
Michael Yang
|
32aec66e6a
add load duration
|
1 éve |
Michael Yang
|
35af37a2cb
session id
|
1 éve |