NLnl·Feb 21, 2023
Obviously fine-tuning is just a special (easier) case of training.
They are working on complete training of large models too: https://github.com/yandex-research/swarm
In Running large language models like ChatGPT on a single GPUyandex-research
swarm
Python2023–2023
yandex-research / swarm
swarm
Official code for "SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient"
2Distinct people
2Mentions
0Explicit recommenders
151GitHub stars
What to know before exploring
What it is
Official code for "SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient"
GitHub
Where it is used
No documented product use has been verified yet.
What people are saying
No recommendation phrase was found in the indexed comments.
No classified warning is attached to this page. Read the discussions before treating silence as approval.
Also in these discussions
No other repository is repeatedly linked in the same discussions.
This week on Hacker News
Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 5:04:57 PM UTC.
No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.
Mentions over time
HN mentions by 6-hour interval2 dated mentions
Observed mentions: 2023-02-15 18:19:47 UTC to 2023-02-21 01:19:10 UTC.
Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.
View mention counts and date ranges
| Period | Start (UTC) | End (exclusive, UTC) | Mentions | Submissions | Comments |
|---|---|---|---|---|---|
| Feb 15, 18:00 | 2023-02-15 18:00:00 UTC | 2023-02-16 00:00:00 UTC | 1 | 0 | 1 |
| Feb 16, 00:00 | 2023-02-16 00:00:00 UTC | 2023-02-16 06:00:00 UTC | 0 | 0 | 0 |
| Feb 16, 06:00 | 2023-02-16 06:00:00 UTC | 2023-02-16 12:00:00 UTC | 0 | 0 | 0 |
| Feb 16, 12:00 | 2023-02-16 12:00:00 UTC | 2023-02-16 18:00:00 UTC | 0 | 0 | 0 |
| Feb 16, 18:00 | 2023-02-16 18:00:00 UTC | 2023-02-17 00:00:00 UTC | 0 | 0 | 0 |
| Feb 17, 00:00 | 2023-02-17 00:00:00 UTC | 2023-02-17 06:00:00 UTC | 0 | 0 | 0 |
| Feb 17, 06:00 | 2023-02-17 06:00:00 UTC | 2023-02-17 12:00:00 UTC | 0 | 0 | 0 |
| Feb 17, 12:00 | 2023-02-17 12:00:00 UTC | 2023-02-17 18:00:00 UTC | 0 | 0 | 0 |
| Feb 17, 18:00 | 2023-02-17 18:00:00 UTC | 2023-02-18 00:00:00 UTC | 0 | 0 | 0 |
| Feb 18, 00:00 | 2023-02-18 00:00:00 UTC | 2023-02-18 06:00:00 UTC | 0 | 0 | 0 |
| Feb 18, 06:00 | 2023-02-18 06:00:00 UTC | 2023-02-18 12:00:00 UTC | 0 | 0 | 0 |
| Feb 18, 12:00 | 2023-02-18 12:00:00 UTC | 2023-02-18 18:00:00 UTC | 0 | 0 | 0 |
| Feb 18, 18:00 | 2023-02-18 18:00:00 UTC | 2023-02-19 00:00:00 UTC | 0 | 0 | 0 |
| Feb 19, 00:00 | 2023-02-19 00:00:00 UTC | 2023-02-19 06:00:00 UTC | 0 | 0 | 0 |
| Feb 19, 06:00 | 2023-02-19 06:00:00 UTC | 2023-02-19 12:00:00 UTC | 0 | 0 | 0 |
| Feb 19, 12:00 | 2023-02-19 12:00:00 UTC | 2023-02-19 18:00:00 UTC | 0 | 0 | 0 |
| Feb 19, 18:00 | 2023-02-19 18:00:00 UTC | 2023-02-20 00:00:00 UTC | 0 | 0 | 0 |
| Feb 20, 00:00 | 2023-02-20 00:00:00 UTC | 2023-02-20 06:00:00 UTC | 0 | 0 | 0 |
| Feb 20, 06:00 | 2023-02-20 06:00:00 UTC | 2023-02-20 12:00:00 UTC | 0 | 0 | 0 |
| Feb 20, 12:00 | 2023-02-20 12:00:00 UTC | 2023-02-20 18:00:00 UTC | 0 | 0 | 0 |
| Feb 20, 18:00 | 2023-02-20 18:00:00 UTC | 2023-02-21 00:00:00 UTC | 0 | 0 | 0 |
| Feb 21, 00:00 | 2023-02-21 00:00:00 UTC | 2023-02-21 06:00:00 UTC | 1 | 0 | 1 |
What Hacker News says
2 sourcesThe BigScience team (a working group of researchers that trained the BLOOM-176B LLM last year) released Petals [0][1] which allows distributed inference and fine-tuning of BLOOM, with the option to pick a custom model + private swarm. SWARM [2][3] is a WIP from yandex and UW that shares some of the same codebase, but is for distributed training.
[0] https://petals.ml/
[1] https://github.com/bigscience-workshop/petals
[2] https://github.com/yandex-research/swarm
[3] https://twitter.com/m_ryabinin/status/1625175933492641814
In Semantic Search with Phoenix, Axon, Bumblebee, and ExFaiss