hpcaitech/EnergonAI

Issues

• How to inference BLOOM-176B by multi-node multi-card？

  #189 opened 2 years ago by vicwer
  2

• 关于示例代码版本落后，无法运行等问题About the example code version backward, can not run and other issues

  #224 opened a year ago by AntyRia
  7

• [bug] The InferenceEngine used in the example for distributed inference cannot be imported.

  #223 opened a year ago by flybird11111
  0

• Support GPT BigCode (bigcode/starcoder, bigcode/gpt_bigcode-santacoder, etc.)

  #218 opened a year ago by liulhdarks
  1

• OPT-125m problem

  #222 opened a year ago by Yummy813
  0

• Location of logs

  #221 opened a year ago by Yummy813
  0

• Docker cannot find the parent image defined in `docker/Dockerfile`

  #220 opened a year ago by Aavache
  0

• Does EnergonAI support accelerated inference for segmenting anything?

  #219 opened a year ago by sanbuphy
  0

• EnergonAI running OPT reasoning example: When encountering a client request, the server is blocked and cannot return the result

  #215 opened 2 years ago by colynhn
  1

• Failed to load pre-trained model weights for OPT_125M

  #217 opened a year ago by zhengmk321
  2

• Where is InferenceEngine definition?

  #216 opened a year ago by yehx1
  0

• question about load model state_dict in multi-gpus

  #213 opened 2 years ago by irasin
  2

• Is there an example of the http client?

  #214 opened 2 years ago by frankxyy
  1

• Concrete doc of this project

  #212 opened 2 years ago by frankxyy
  1

• can't find server.sh

  #160 opened 2 years ago by zhangyilalala
  3

• _pickle.UnpicklingError: invalid load key, '{'.

  #211 opened 2 years ago by RundongCao
  4

• Why does it unreadable generated by OPT-30B inferring with EnergonAI

  #187 opened 2 years ago by ericxsun
  11

• CUDA error: no kernel image is available for execution on the device

  #152 opened 2 years ago by KastanDay
  3

• OPT inference

  #198 opened 2 years ago by Joanna-0421
  2

• Where is InferenceEngine definition?

  #210 opened 2 years ago by liujuncn
  2

• an error caused by running the example of the opt

  #206 opened 2 years ago by LemonSqi
  4

• Cannot run opt 125m examples with latest energonai docker images

  #186 opened 2 years ago by zhanghaoie
  2

• How to use dynamic batch features

  #199 opened 2 years ago by hudengjunai
  1

• OPT demo TEST

  #203 opened 2 years ago by Batizhao8899
  2

• fail to install EnergonAI

  #204 opened 2 years ago by NewDriverLee
  3

• Is there any examples of using offload feature in GPT/BLOOM/OPT inference?

  #209 opened 2 years ago by YJHMITWEB
  1

• miss cache error when pose generation opt

  #201 opened 2 years ago by tycallen
  2

• failure to compile energonai by the command : python setup.py build

  #205 opened 2 years ago by LemonSqi
  1

• Doesn't run gpt reference?

  #196 opened 2 years ago by YuchengWang
  1

• Does not support Cuda 10.2 ?

  #193 opened 2 years ago by 0-1CxH
  1

• Not compatible with the latest version of transformers? (4.26.1)

  #192 opened 2 years ago by skiingpacman
  2

• Can not start the Bloom server

  #191 opened 2 years ago by SAI990323
  3

• Maybe you should add license for using OneFlow's LayerNorm Kernel implement？

  #194 opened 2 years ago by MARD1NO
  2

• Failed to load OPT-30B checkpoint

  #183 opened 2 years ago by ericxsun
  2

• Support OPT-IML model

  #184 opened 2 years ago by ericxsun
  1

• Detected RRef Leaks during shutdown, empty pipe, tests_engine failed

  #182 opened 2 years ago by nostalgicimp
  1

• trpc.rpc_sync consumed most time

  #175 opened 2 years ago by fanlongbd
  0

• RuntimeError('FusedLayerNormAffineFunction requires cuda extensions')

  #174 opened 2 years ago by sori424
  0

• torch.load() hangs indefinitely when reading OPT pre-trained model weights

  #159 opened 2 years ago by larry-fuy
  1

• need guidelines on converting OPT-17B checkpoint

  #161 opened 2 years ago by gulzainali98
  0

• does EnergonAI support gpt model with int8 quantitation in model parallel?

  #158 opened 2 years ago by dearowen
  1

• [RFC] Async engine and pipeline based on RPC

  #151 opened 2 years ago by ver217
  1

• num_beams for beam search

  #138 opened 2 years ago by shammmmmless
  1

• inference of pre-trained model

  #125 opened 2 years ago by Emerald01
  1

• Remove hard code directory path

  #110 opened 2 years ago by feifeibear
  0

• Provide a docker service

  #106 opened 2 years ago by feifeibear
  0

• OPT inference generate example

  #102 opened 2 years ago by virgulvirgul
  1

• Missing energonai_linear_func in setup.py

  #100 opened 2 years ago by xgreat8
  1

• Connection refused on docker exposed port

  #99 opened 2 years ago by xgreat8
  1

• [Feature]: Automatic Pipeline Parallelism

  #89 opened 2 years ago by dujiangsu
  0