# Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL

DevFeed: [Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL](<https://devfeed.tech/articles/shipping-a-trillion-parameters-with-a-hub-bucket-delta-weight-sync-in-trl-7166.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/delta-weight-sync>)

Author: Amine Dirhoussi; Quentin Gallouédec; Kashif Rasul; Lewis Tunstall; Edward Beeching; Albert Villanova del Moral; Leandro von Werra; Sergio Paniego

Published: 2026-05-27T00:00:00Z

Content type: article

Language: en

Sources: [Hugging Face - Blog](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [AI, ML & Data Engineering](<https://devfeed.tech/topics/ai-ml-data-engineering.md>), [NCCL](<https://devfeed.tech/topics/nccl.md>)

Tags: [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [compute](<https://devfeed.tech/tags/compute.md>), [hub](<https://devfeed.tech/tags/hub.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [inference](<https://devfeed.tech/tags/inference.md>), [llm](<https://devfeed.tech/tags/llm.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [nccl](<https://devfeed.tech/tags/nccl.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [payload](<https://devfeed.tech/tags/payload.md>), [policy](<https://devfeed.tech/tags/policy.md>), [rl](<https://devfeed.tech/tags/rl.md>), [space](<https://devfeed.tech/tags/space.md>), [storage](<https://devfeed.tech/tags/storage.md>), [sync](<https://devfeed.tech/tags/sync.md>), [train](<https://devfeed.tech/tags/train.md>), [training](<https://devfeed.tech/tags/training.md>), [trl](<https://devfeed.tech/tags/trl.md>), [update](<https://devfeed.tech/tags/update.md>), [vllm](<https://devfeed.tech/tags/vllm.md>)

## AI overview

The article describes delta weight synchronization for asynchronous reinforcement-learning training. A TRL change stores only modified model weights in sparse safetensors files and lets vLLM fetch them from a Hugging Face bucket, reducing transfer payloads and enabling disaggregated training without a shared cluster.

## Source excerpt

We're on a journey to advance and democratize artificial intelligence through open source and open science.