Stefan Gligorijevic
|
5dbd262033
feat: Mirostat v2 (#69)
|
1 年之前 |
AlpinDale
|
f393dc2af1
fix: broken GPTQ layer
|
1 年之前 |
AlpinDale
|
3bf6197afb
fix: prompt processing delay introduced by #66 (#71)
|
1 年之前 |
AlpinDale
|
9df91fe863
bump version to 0.3.6
|
1 年之前 |
AlpinDale
|
2b42a1ada2
bump the version to 0.3.5
|
1 年之前 |
AlpinDale
|
c55c8f7bd8
update readme
|
1 年之前 |
AlpinDale
|
380206038e
fix: change the timing of logit sorting (#66)
|
1 年之前 |
AlpinDale
|
bdad759503
feat: YaRN context window extension support (#67)
|
1 年之前 |
AlpinDale
|
f04588203e
feat: mistral AWQ support and file blacklisting
|
1 年之前 |
AlpinDale
|
7572e1dd59
overflow in AWQ GEMM kernel
|
1 年之前 |
AlpinDale
|
c1fa7e8567
chore: fix datatype check (#65)
|
1 年之前 |
AlpinDale
|
a6a4220fa6
feat: refactor megatron and quants (#57)
|
1 年之前 |
AlpinDale
|
9a9e59b871
update readme with new instructions
|
1 年之前 |
g4rg
|
16bf6b61a3
fix: requests stalling in KAI non-streaming endpoint (#46)
|
1 年之前 |
LitreallyNone
|
b526a7b3bc
Update requirements.txt (#58)
|
1 年之前 |
AlpinDale
|
2e70a6d5ed
chore: allow the user to specify install method (#56)
|
1 年之前 |
official-elinas
|
46e472062a
chore: make NVCC work for different versions (#55)
|
1 年之前 |
AlpinDale
|
6682ede3de
fix: clean up API servers
|
1 年之前 |
henk717
|
0b2b62fe96
Micromamba Runtime (#54)
|
1 年之前 |
AlpinDale
|
1e294e1bfa
include klite UI in the build
|
1 年之前 |
AlpinDale
|
9f7a0e3ecb
feat: AWQ support for Turing GPUs (#53)
|
1 年之前 |
AlpinDale
|
1874afabce
readme: fix rope instructions
|
1 年之前 |
AlpinDale
|
044251018e
chore: update readme and the tests
|
1 年之前 |
AlpinDale
|
b7918ad45f
fix: attention kernel attribute (#52)
|
1 年之前 |
Sen
|
4a93dcbbe3
chore: add CORS middleware (#51)
|
1 年之前 |
g4rg
|
a522960f6e
fix: more KAI parameter adaptations (#45)
|
1 年之前 |
g4rg
|
ccb6db3d6a
fix: add kcpp /generate/check stub (#47)
|
1 年之前 |
AlpinDale
|
c5869e0b62
feat: add docker image
|
1 年之前 |
AlpinDale
|
a6fb3d54e0
bump version to v0.3.4
|
1 年之前 |
LostRuins
|
6d4bdf374c
feat: add Kobold Lite UI (#42)
|
1 年之前 |