Skip to content

Issues: pytorch/torchtune

v0.7.0 tracker
#2538 opened Mar 30, 2025 by joecummings
Open
Testing tracker
#1890 opened Oct 23, 2024 by felipemello1
Open
Beta
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

310 Open 727 Closed
310 Open 727 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
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
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Qwen2/2.5/3 max_seq_len and EOS Masking
#2792 opened Jun 5, 2025 by haydn-jones
is_multimodal_input not defined in generate_v2.py good first issue Good for newcomers
#2790 opened Jun 5, 2025 by mreso
1
Generation in full_finetune_distributed recipe
#2786 opened Jun 5, 2025 by FSadrieh
@SalmanMohammadi
2
Proposal: reuse methods in recipes
#2779 opened Jun 2, 2025 by Andrei-Aksionov
2
Torchtune dataset cache memory error despite providing dataset file path
#2775 opened May 30, 2025 by zabir110
@ebsmothers
2
Perf / accuracy metrics comparison with Nemo for SFT / reasoning distillation scenario
#2760 opened May 22, 2025 by vadimkantorov
5
torchtune fails to start when the checkpointer dir does not contain the initial model weights enhancement New feature or request
#2759 opened May 22, 2025 by vadimkantorov
2
chat_dataset(...) should use newer masking_strategy arg
#2758 opened May 22, 2025 by vadimkantorov
2
torchao dependency not listed
#2752 opened May 20, 2025 by cshimmin
2
Integrate new OptimizerInBackward in rest of torchtune codebase
#2750 opened May 19, 2025 by joecummings
4 tasks
2
I'm experimenting unexpected loss values when running old recipes on the same datasets bug Something isn't working
#2746 opened May 17, 2025 by nouranali
@joecummings
12
async GRPO discarding data and wandb not logging rewards triage review This issue should be discussed in weekly review
#2739 opened May 15, 2025 by salman-abdullah
@joecummings
3
Bidirectional truncation in Llama4. best practice Things we should be doing but aren't community help wanted We would love the community's help completing this issue
#2735 opened May 15, 2025 by krammnic
5
Feature request: support different input/output formats in the same recipe
#2732 opened May 14, 2025 by andrewor14
1
[RFC] Automatically download models as part of tune run discussion Start a discussion
#2731 opened May 14, 2025 by ebsmothers
Use LRScheduler.load_state_dict() best practice Things we should be doing but aren't community help wanted We would love the community's help completing this issue
#2730 opened May 13, 2025 by joecummings
1
2
Integrate Muon optimizer community help wanted We would love the community's help completing this issue enhancement New feature or request
#2725 opened May 13, 2025 by joecummings
@Saurabh750
2
Migrate typing to PEP 585 and PEP 604 best practice Things we should be doing but aren't
#2715 opened May 9, 2025 by joecummings
3
Drop Python 3.9 Support discussion Start a discussion
#2713 opened May 9, 2025 by joecummings
Activation offloading and Llama4
#2709 opened May 8, 2025 by ebsmothers
1
Integrate Hugging Face tokenizers with torchtune community help wanted We would love the community's help completing this issue high-priority
#2706 opened May 8, 2025 by ebsmothers
1
Open call for projects built with torchtune! 🚀 community help wanted We would love the community's help completing this issue
#2701 opened May 8, 2025 by joecummings
Log config to output only on rank zero best practice Things we should be doing but aren't community help wanted We would love the community's help completing this issue good first issue Good for newcomers
#2700 opened May 8, 2025 by joecummings
Add support for Qwen 2.5 VL 7B community help wanted We would love the community's help completing this issue enhancement New feature or request
#2699 opened May 8, 2025 by joecummings
4
Add a knowledge distillation full finetune distributed recipe community help wanted We would love the community's help completing this issue enhancement New feature or request
#2695 opened May 7, 2025 by joecummings
1
ProTip! Type g i on any issue or pull request to go back to the issue listing page.