coreweave/nccl-tests issues and pull requests

#42 - Fix Grammar in README.md

Pull Request - State: open - Opened by mtnsteeptea 2 months ago

#41 - feat: Add CUDA 12.6 builds, update HPC-X & NCCL

Pull Request - State: closed - Opened by Eta0 3 months ago
Labels: enhancement

#40 - chore: Update example to latest image

Pull Request - State: closed - Opened by NavarrePratt 6 months ago

#39 - How use nvidia sharp for training job

Issue - State: open - Opened by Lzhang-hub 8 months ago - 1 comment

#38 - feat: Update HPC-X to v2.19

Pull Request - State: closed - Opened by Eta0 8 months ago - 11 comments
Labels: enhancement

#37 - feat: Update CUDA 12.4 builds to 12.4.1 with cuDNN

Pull Request - State: closed - Opened by Eta0 9 months ago
Labels: enhancement

#36 - build: Label the `base` build stage

Pull Request - State: closed - Opened by Eta0 10 months ago - 6 comments
Labels: bug

#35 - feat: Build NCCL from source for more up-to-date versions

Pull Request - State: open - Opened by Eta0 10 months ago
Labels: enhancement

#34 - fix: Build GDRCopy from source on Ubuntu 20.04 as well

Pull Request - State: closed - Opened by Eta0 10 months ago

#33 - feat: Add CUDA 12.4 builds, update NCCL

Pull Request - State: closed - Opened by Eta0 10 months ago
Labels: enhancement

#32 - docs: Update README with new image tags

Pull Request - State: closed - Opened by Eta0 10 months ago
Labels: documentation

#31 - feat: Update NCCL, CUDA, cuDNN, and HPC-X

Pull Request - State: closed - Opened by Eta0 11 months ago - 9 comments
Labels: enhancement

#30 - ci: Downgrade `ubuntu22` + `cu120`'s `nccl-version` to `2.18.5-1`

Pull Request - State: closed - Opened by Eta0 about 1 year ago - 3 comments
Labels: bug

#29 - feat: Restore `ubuntu22.04` builds

Pull Request - State: closed - Opened by Eta0 about 1 year ago
Labels: enhancement

#28 - refactor: Consolidate HPC-X build arguments

Pull Request - State: closed - Opened by Eta0 about 1 year ago
Labels: enhancement

#27 - build: Add disabled CUDA 12.3, add `--allow-downgrades`.

Pull Request - State: closed - Opened by wbrown about 1 year ago - 16 comments

#26 - build: Update to NCCL v2.18.5 & CUDA 12.2.2

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 5 comments
Labels: enhancement

#25 - build: Update HPC-X to v2.16 for CUDA 12 builds

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 7 comments
Labels: enhancement

#25 - build: Update HPC-X to v2.16 for CUDA 12 builds

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 7 comments
Labels: enhancement

#24 - feat: Allow specifying a custom cuDNN version

Pull Request - State: open - Opened by Eta0 over 1 year ago - 8 comments
Labels: enhancement

#23 - NCCL v2.18.3 & CUDA 12.2

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 4 comments
Labels: enhancement

#23 - NCCL v2.18.3 & CUDA 12.2

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 4 comments
Labels: enhancement

#22 - fix: newer version of perftest with improved min bw sampling

Pull Request - State: closed - Opened by antgun42 over 1 year ago

#22 - fix: newer version of perftest with improved min bw sampling

Pull Request - State: closed - Opened by antgun42 over 1 year ago

#21 - feat: change perftest to cw fork with min bw

Pull Request - State: closed - Opened by antgun42 over 1 year ago - 4 comments

#21 - feat: change perftest to cw fork with min bw

Pull Request - State: closed - Opened by antgun42 over 1 year ago - 4 comments

#20 - feat: Add slurm examples

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago

#20 - feat: Add slurm examples

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago

#19 - Add slurm sbatch jobs

Issue - State: closed - Opened by salanki over 1 year ago

#19 - Add slurm sbatch jobs

Issue - State: closed - Opened by salanki over 1 year ago

#18 - fix: Stop disabling LL128 for the H100s

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago

#18 - fix: Stop disabling LL128 for the H100s

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago

#17 - feat: Add an H100 example

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago

#17 - feat: Add an H100 example

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago

#16 - feat: Preserve HPC-X environment variables on `ssh` and `sudo`

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 2 comments

#16 - feat: Preserve HPC-X environment variables on `ssh` and `sudo`

Pull Request - State: closed - Opened by Eta0 over 1 year ago - 2 comments

#15 - feat: Add 12.1 image and update 12.* versions

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago - 12 comments

#15 - feat: Add 12.1 image and update 12.* versions

Pull Request - State: closed - Opened by NavarrePratt over 1 year ago - 12 comments

#14 - Add NCCL v2.18.0-1 builds for CUDA 11.8 & 12.0

Pull Request - State: closed - Opened by Eta0 almost 2 years ago - 6 comments
Labels: enhancement

#14 - Add NCCL v2.18.0-1 builds for CUDA 11.8 & 12.0

Pull Request - State: closed - Opened by Eta0 almost 2 years ago - 6 comments
Labels: enhancement

#13 - Base builds off of cudnn8-devel images

Pull Request - State: closed - Opened by Eta0 almost 2 years ago - 1 comment

#13 - Base builds off of cudnn8-devel images

Pull Request - State: closed - Opened by Eta0 almost 2 years ago - 1 comment

#12 - Fix HPC-X library path prefixes

Pull Request - State: closed - Opened by Eta0 almost 2 years ago
Labels: bug

#12 - Fix HPC-X library path prefixes

Pull Request - State: closed - Opened by Eta0 almost 2 years ago
Labels: bug

#11 - Fix invalid ENV definition when setting UCX_VFS_ENABLE=no

Pull Request - State: closed - Opened by Eta0 almost 2 years ago

#11 - Fix invalid ENV definition when setting UCX_VFS_ENABLE=no

Pull Request - State: closed - Opened by Eta0 almost 2 years ago

#10 - Fix CUDA 12.0.1 image tag in README

Pull Request - State: closed - Opened by Eta0 almost 2 years ago - 1 comment

#10 - Fix CUDA 12.0.1 image tag in README

Pull Request - State: closed - Opened by Eta0 almost 2 years ago - 1 comment

#9 - Fix CI variable for custom NCCL version

Pull Request - State: closed - Opened by salanki almost 2 years ago

#8 - CUDA 12.0.1

Pull Request - State: closed - Opened by salanki almost 2 years ago - 3 comments

#8 - CUDA 12.0.1

Pull Request - State: closed - Opened by salanki almost 2 years ago - 3 comments

#7 - CUDA 12.0.1

Pull Request - State: closed - Opened by salanki almost 2 years ago

#7 - CUDA 12.0.1

Pull Request - State: closed - Opened by salanki almost 2 years ago

#6 - feat(chore): Ensure mkdir has -p flag

Pull Request - State: closed - Opened by rtalaricw about 2 years ago

#6 - feat(chore): Ensure mkdir has -p flag

Pull Request - State: closed - Opened by rtalaricw about 2 years ago

#5 - ci(github): Add image builder workflows

Pull Request - State: closed - Opened by todie about 2 years ago - 8 comments

#5 - ci(github): Add image builder workflows

Pull Request - State: closed - Opened by todie about 2 years ago - 8 comments

#4 - docs(readme): Fix typos

Pull Request - State: closed - Opened by m1tttt4 over 2 years ago