issues
search
hpcaitech
/
EnergonAI
Large-scale model inference.
Apache License 2.0
630
stars
90
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
trpc.rpc_sync consumed most time
#175
fanlongbd
opened
1 year ago
0
RuntimeError('FusedLayerNormAffineFunction requires cuda extensions')
#174
sori424
opened
1 year ago
0
Update LICENSE
#173
binmakeswell
closed
1 year ago
0
Bloom int8 mode prototype
#172
CsRic
closed
1 year ago
0
bloom fp16 + tp server
#171
CsRic
closed
1 year ago
0
add shard init for tp
#170
CsRic
closed
1 year ago
0
TP shared cpu model
#169
CsRic
closed
1 year ago
0
update batch process
#168
CsRic
closed
1 year ago
0
[doc] polish readme
#167
feifeibear
closed
1 year ago
0
add TP for bloom
#166
feifeibear
closed
1 year ago
0
[bloom]update batch process
#165
CsRic
closed
1 year ago
0
bloom service
#164
CsRic
closed
1 year ago
0
[docs] add guide to pre-processing opt-175b weights
#163
ver217
closed
1 year ago
0
[opt] add opt-6.7b and add fastapi server
#162
ver217
closed
1 year ago
0
need guidelines on converting OPT-17B checkpoint
#161
gulzainali98
closed
1 year ago
0
can't find server.sh
#160
zhangyilalala
opened
1 year ago
3
torch.load() hangs indefinitely when reading OPT pre-trained model weights
#159
larry-fuy
opened
1 year ago
1
does EnergonAI support gpt model with int8 quantitation in model parallel?
#158
dearowen
opened
2 years ago
1
[engine] Async engine and pipeline based on RPC
#157
ver217
closed
1 year ago
1
[setup] add version control
#156
ver217
closed
2 years ago
0
[utils] fix checkpointing import
#155
ver217
closed
2 years ago
0
[utils] remove useless utils
#154
ver217
closed
2 years ago
0
[logging] remove logging module
#153
ver217
closed
2 years ago
0
CUDA error: no kernel image is available for execution on the device
#152
KastanDay
opened
2 years ago
3
[RFC] Async engine and pipeline based on RPC
#151
ver217
closed
1 year ago
1
prometheus
#150
feifeibear
closed
2 years ago
0
Add prometheus endpoint for opt_server.py
#149
ofey404
closed
2 years ago
1
update readme
#148
dujiangsu
closed
2 years ago
0
[nn] replace energonai.nn with colossalai.nn
#147
ver217
closed
2 years ago
0
[opt] add queue size option
#146
ver217
closed
2 years ago
0
[opt] add timeout option
#145
ver217
closed
2 years ago
0
update readme
#144
dujiangsu
closed
2 years ago
0
[docker] rm source code after pip install
#143
feifeibear
closed
2 years ago
0
[opt] refactor server
#142
ver217
closed
2 years ago
0
processing 66b ckpt
#141
dujiangsu
closed
2 years ago
0
[opt] allow disabling multi procs loading
#140
ver217
closed
2 years ago
0
[opt] fit opt-175b
#139
ver217
closed
2 years ago
0
num_beams for beam search
#138
shammmmmless
opened
2 years ago
1
66B model load checkpoint
#137
dujiangsu
closed
2 years ago
0
fit the opt_66B
#136
dujiangsu
closed
2 years ago
0
[opt] add cache and modify api
#135
ver217
closed
2 years ago
0
improve the cache implementation
#134
dujiangsu
closed
2 years ago
0
[opt] executor update making batch policy
#133
ver217
closed
2 years ago
0
[opt] add data validator and cors middleware
#132
ver217
closed
2 years ago
0
add a flag for disable cache
#131
dujiangsu
closed
2 years ago
0
Generation model feature: add cache for removing the repeated computation in the loop.
#130
dujiangsu
closed
2 years ago
0
[hotfix] fix worker server shutdown
#129
ver217
closed
2 years ago
0
[opt] remove useless api
#128
ver217
closed
2 years ago
0
[opt] add 175b model
#127
ver217
closed
2 years ago
0
[opt] add left padding
#126
ver217
closed
2 years ago
0
Previous
Next