osdc/hf-cache: scripts/hf-cache-seed.py to seed models into bucket(s)#833
Open
huydhn wants to merge 18 commits into
Open
osdc/hf-cache: scripts/hf-cache-seed.py to seed models into bucket(s)#833huydhn wants to merge 18 commits into
huydhn wants to merge 18 commits into
Conversation
This was referenced Jun 24, 2026
tofu plan — arc-cbr-production✅ Plan succeeded · commit Plan output |
tofu plan — meta-prod-aws-ue1✅ Plan succeeded · commit Plan output |
tofu plan — lf-prod-aws-ue2✅ Plan succeeded · commit Plan output |
huydhn
added a commit
that referenced
this pull request
Jun 25, 2026
Adds a local one-off recipe to seed a model into a cluster's HF cache S3 bucket: downloads with huggingface_hub and `aws s3 sync`s to s3://pytorch-hf-model-cache-<cluster_id>/hub using your AWS creds (not the OIDC CI role). Default model Qwen/Qwen2.5-7B-Instruct; set HF_TOKEN for gated models. Used to populate the large model that test-hf-cache-large-read reads. ghstack-source-id: 6d009e9 Pull-Request: #833
tofu plan — arc-cbr-production-uw1❌ Plan failed · commit Plan output |
huydhn
added a commit
that referenced
this pull request
Jun 25, 2026
Adds a local one-off recipe to seed a model into a cluster's HF cache S3 bucket: downloads with huggingface_hub and `aws s3 sync`s to s3://pytorch-hf-model-cache-<cluster_id>/hub using your AWS creds (not the OIDC CI role). Default model Qwen/Qwen2.5-7B-Instruct; set HF_TOKEN for gated models. Used to populate the large model that test-hf-cache-large-read reads. ghstack-source-id: 6d009e9 Pull-Request: #833
huydhn
added a commit
that referenced
this pull request
Jun 25, 2026
Adds a local one-off recipe to seed a model into a cluster's HF cache S3 bucket: downloads with huggingface_hub and `aws s3 sync`s to s3://pytorch-hf-model-cache-<cluster_id>/hub using your AWS creds (not the OIDC CI role). Default model Qwen/Qwen2.5-7B-Instruct; set HF_TOKEN for gated models. Used to populate the large model that test-hf-cache-large-read reads. ghstack-source-id: e0a82f4 Pull-Request: #833
huydhn
added a commit
that referenced
this pull request
Jun 25, 2026
Standalone seeding tool under osdc/scripts (moved out of the justfile). Downloads each HF model once locally, then `aws s3 sync`s it to one or more clusters' per-region buckets pytorch-hf-model-cache-<cluster_id>, or --all clusters that enable the hf-cache module (synced in parallel). Writes straight to S3 with the operator's AWS creds — independent of the cluster mount. Run once per model. uv run scripts/hf-cache-seed.py -c meta-staging-aws-ue1 Qwen/Qwen2.5-7B-Instruct uv run scripts/hf-cache-seed.py --all Qwen/Qwen2.5-7B-Instruct ghstack-source-id: 8866009 Pull-Request: #833
huydhn
added a commit
that referenced
this pull request
Jun 25, 2026
Standalone seeding tool under osdc/scripts (moved out of the justfile). Downloads each HF model once locally, then `aws s3 sync`s it to one or more clusters' per-region buckets pytorch-hf-model-cache-<cluster_id>, or --all clusters that enable the hf-cache module (synced in parallel). Writes straight to S3 with the operator's AWS creds — independent of the cluster mount. Run once per model. uv run scripts/hf-cache-seed.py -c meta-staging-aws-ue1 Qwen/Qwen2.5-7B-Instruct uv run scripts/hf-cache-seed.py --all Qwen/Qwen2.5-7B-Instruct ghstack-source-id: a185035 Pull-Request: #833
huydhn
added a commit
that referenced
this pull request
Jun 25, 2026
Standalone seeding tool under osdc/scripts (moved out of the justfile). Downloads each HF model once locally, then `aws s3 sync`s it to one or more clusters' per-region buckets pytorch-hf-model-cache-<cluster_id>, or --all clusters that enable the hf-cache module (synced in parallel). Writes straight to S3 with the operator's AWS creds — independent of the cluster mount. Run once per model. uv run scripts/hf-cache-seed.py -c meta-staging-aws-ue1 Qwen/Qwen2.5-7B-Instruct uv run scripts/hf-cache-seed.py --all Qwen/Qwen2.5-7B-Instruct ghstack-source-id: 3769c61 Pull-Request: #833
huydhn
added a commit
that referenced
this pull request
Jun 26, 2026
Standalone seeding tool under osdc/scripts (moved out of the justfile). Downloads each HF model once locally, then `aws s3 sync`s it to one or more clusters' per-region buckets pytorch-hf-model-cache-<cluster_id>, or --all clusters that enable the hf-cache module (synced in parallel). Writes straight to S3 with the operator's AWS creds — independent of the cluster mount. Run once per model. uv run scripts/hf-cache-seed.py -c meta-staging-aws-ue1 Qwen/Qwen2.5-7B-Instruct uv run scripts/hf-cache-seed.py --all Qwen/Qwen2.5-7B-Instruct ghstack-source-id: 471693c Pull-Request: #833
tofu plan — meta-prod-aws-uw1✅ Plan succeeded · commit Plan output |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
Standalone seeding tool under osdc/scripts (moved out of the justfile). Downloads
each HF model once locally, then
aws s3 syncs it to one or more clusters'per-region buckets pytorch-hf-model-cache-<cluster_id>, or --all clusters that
enable the hf-cache module (synced in parallel). Writes straight to S3 with the
operator's AWS creds — independent of the cluster mount. Run once per model.
uv run scripts/hf-cache-seed.py -c meta-staging-aws-ue1 Qwen/Qwen2.5-7B-Instruct
uv run scripts/hf-cache-seed.py --all Qwen/Qwen2.5-7B-Instruct