alpa-projects/alpa

Issues

• Working Environment for Alpa?

  #975 opened 2 months ago by man2machine
  0

• Fail to run alpa test

  #937 opened 2 years ago by gaow0007
  2

• How to build debug-version Alpa-modified jaxlib

  #959 opened a year ago by caixiiaoyang
  3

• python3 -m alpa.test_install error

  #974 opened 6 months ago by sandy99-w
  0

• Check Installation failled

  #973 opened a year ago by tangyuelm
  1

• Install Alpa without GPU

  #943 opened a year ago by AlbertZhangHIT
  1

• F external/org_tensorflow/tensorflow/compiler/xla/python/xla.cc:227] Check failed: stream_device != nullptr (0 vs. nullptr)

  #901 opened 2 years ago by chaokunyang
  1

• [Bug] Segment fault when using alpa to parallelize llama with jax 0.4.6 environment

  #970 opened a year ago by zigzagcai
  2

• Is Alpa still under active development?

  #947 opened a year ago by abatilo
  3

• How to profile Alpa models and get the trace

  #971 opened a year ago by zigzagcai
  1

• Check failed: operand_dim < ins->operand(0)->shape().rank() (2 vs. 2)Does not support this kind of Gather.

  #958 opened a year ago by caixiiaoyang
  2

• Ray spill out of disk error when using alpa to auto-parallelize llama

  #969 opened a year ago by zigzagcai
  2

• Problem in building Alpa-modified Jaxlib.

  #956 opened a year ago by Fonsifa
  5

• Why did you choose ray instead of using torch distributed?

  #968 opened a year ago by HongLouyemeng
  2

• Any solution to support llama2 finetune?

  #966 opened a year ago by LeiWang1999
  0

• A question about file /alpa/benchmark/gen_serving_database.py

  #965 opened a year ago by God68
  0

• Error about python3 -m alpa.test_install

  #964 opened a year ago by God68
  0

• How to use Alpa to serve BERT models

  #962 opened a year ago by Jessssssie99
  0

• Unsupported parallel mode in shard-only auto perf test: load_solution

  #961 opened a year ago by LeiWang1999
  0

• when i check installation by running python3 -m alpa.test_install,AssertionError happend

  #960 opened a year ago by Ghiad
  6

• Does not support this kind of Gather

  #941 opened a year ago by bojone
  2

• IndexError: `InlinedVector::at(size_type) const` failed bounds check

  #957 opened a year ago by caixiiaoyang
  0

• Failed With Installation Check

  #918 opened 2 years ago by szxiangjn
  3

• Unable to use pipeline parallelism with multi-node meshes

  #953 opened a year ago by Edwinhe03
  1

• Question abuot licence / usage

  #955 opened a year ago by EnricoBeltramo
  1

• PLS, a paper related question I want to ask

  #954 opened a year ago by KylinC
  1

• cupy package mismatches with CUDA version in the docs

  #950 opened a year ago by serach24
  2

• Will alpa support jax 0.4.x and cuda 12.x?

  #949 opened a year ago by horse6
  0

• PLS! why each device mesh's memory is always full?

  #948 opened a year ago by KylinC
  1

• Fail to run alpa.test_install

  #946 opened a year ago by weinanliu
  2

• Support PyTorch XLA as a frontend

  #945 opened a year ago by Kelvin-Ng
  3

• Check failed: !info.content.empty()

  #944 opened a year ago by MoFHeka
  0

• support linux aarch64

  #942 opened a year ago by AlbertZhangHIT
  0

• Building alpa/jaxlib for TPU

  #924 opened a year ago by Lime-Cakes
  2

• 4090 not support

  #940 opened a year ago by cery999
  1

• Will OPT-350m be supported

  #939 opened 2 years ago by Cascol-Chen
  0

• Question about auto_sharding.py

  #936 opened 2 years ago by kevinsouthxu
  0

• Question concerning training

  #930 opened 2 years ago by nferreira
  0

• Runtime link broken in docs

  #929 opened 2 years ago by jeromeku
  0

• ImportError: cannot import name 'FrozenDict' from 'jax.experimental.maps'

  #922 opened 2 years ago by carlosgmartin
  4

• Install llm_serving without having to clone the repo

  #919 opened 2 years ago by gianlucadetommaso
  0

• feat(build): Update alpa's build of XLA and PyTorch

  #916 opened 2 years ago by jon-chuang
  0

• Specify Ray instance port

  #913 opened 2 years ago by akhmedsakip
  1

• Why do not use the latest `jaxlib` and `jax`?

  #911 opened 2 years ago by chaoming0625
  2

• AttributeError: 'DistributedArray' object has no attribute 'uuid'

  #904 opened 2 years ago by chaokunyang
  0

• Rendezvous crashes while trying to access already killed NCCLUniqueIDStore

  #907 opened 2 years ago by gjoliver
  0

• [Discussion]On the pillar of alpa's two-level hierarchical space of parallelism

  #906 opened 2 years ago by GHGmc2
  6

• Sync global_env.global_config across Ray cluster.

  #898 opened 2 years ago by gjoliver
  1

• Jax dispatch fail at `TypeError: No device_put handler ` when passing DistributedArray

  #900 opened 2 years ago by chaokunyang
  0

• alpa.test_install error

  #897 opened 2 years ago by vectercyg
  3