Skip to content

Pull requests: AMD-AGI/Primus

New pull request New
18 Open 483 Closed
18 Open 483 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

ci(benchmark): add more model for single node
#509 opened Jan 23, 2026 by HuangWei-95 Loading…
tests: run megatron trainer UT via primus-cli direct (instead of run_pretrain.sh)
#508 opened Jan 23, 2026 by Xiaoming-AMD Loading…
2
docs(cli): add blog on unified training entry with Primus CLI
#505 opened Jan 23, 2026 by Xiaoming-AMD Loading…
1
fix(yaml): remove deprecated turbo option from llama4 config
#502 opened Jan 21, 2026 by RuibinCheung Loading…
2
Fixes RCCL all‑to‑all benchmark to use torch.distributed.group.WORLD only after init_process_group().
#501 opened Jan 21, 2026 by Vivicai1005 Loading…
1
Feature/backends/megatron mlflow status
#498 opened Jan 20, 2026 by mvstrauss Loading…
11
Feat/combined test
#495 opened Jan 15, 2026 by gphuang • Draft
1
[WIP] Multinode projection with different parallelization strategies when single node is benchmarked
#492 opened Jan 14, 2026 by araina-amd • Draft
52
Training Benchmark
#464 opened Dec 29, 2025 by wenxie-amd • Draft
20
feat: Add MLflow artifact upload for traces and logs
#440 opened Dec 18, 2025 by gphuang Loading…
82
feat: Add TraceLens integration for trace analysis with MLflow upload
#439 opened Dec 18, 2025 by gphuang Loading…
216
[Feature] Mamba Trainer Integration
#348 opened Dec 10, 2025 by clairesonglee • Draft
4
Feature/backends/megatron mlflow env
#345 opened Dec 10, 2025 by mvstrauss Loading…
41
Feat/fix mlflow multinode invalid metric log
#344 opened Dec 9, 2025 by olehtika Loading…
6
fix(megatron): handle 2-element tensor in loss reducer
#340 opened Dec 5, 2025 by dzautner Loading…
megatron: remove useless parameters from base configuration
#333 opened Dec 3, 2025 by takaswie Loading…
Add GitHub Actions workflow for Llama3.1 8B training
#297 opened Nov 20, 2025 by kylasa Loading…
12
[WIP] Methods to aggregate figures of merit
#207 opened Sep 22, 2025 by jaredAMD Loading…
1
ProTip! Follow long discussions with comments:>50.