-
Notifications
You must be signed in to change notification settings - Fork 127
Issues: fla-org/flash-linear-attention
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[Bug] Inference Error in Sliding Window Attention Layer Due to Out-of-Bounds Indexing
bug
Something isn't working
#223
opened Mar 11, 2025 by
Lynn-020809
2 tasks done
[Bug] RuntimeError: Triton Error [CUDA]: context is destroyed
bug
Something isn't working
#222
opened Mar 11, 2025 by
Triang-jyed-driung
2 tasks done
[Feature Request] Customize initialization / Add a switch for turning off FLA's initialization
enhancement
New feature or request
#220
opened Mar 10, 2025 by
Triang-jyed-driung
[Feature Request] Allow Q K has different sequence len
enhancement
New feature or request
#219
opened Mar 10, 2025 by
vanhowe
[RFC] Support context parallelism
enhancement
New feature or request
#217
opened Mar 7, 2025 by
sustcsonglin
[Feature Request] Enhancing Unit Testing for FLA in the Context of Active Development and Diverse GPU Compatibility
enhancement
New feature or request
todo
To be implemented
#209
opened Mar 1, 2025 by
uniartisan
[RFC] Increase computation intensity for certain kernels
enhancement
New feature or request
#190
opened Feb 17, 2025 by
sustcsonglin
[RFC] enhanced evaluation support
enhancement
New feature or request
#184
opened Feb 13, 2025 by
sustcsonglin
[RFC] Fuse elementwise operations in RWKV layers
enhancement
New feature or request
#165
opened Feb 5, 2025 by
sustcsonglin
[RFC] Support more hybrid patterns
enhancement
New feature or request
urgent
#153
opened Feb 1, 2025 by
sustcsonglin
[RFC] Implement model-specific 4d parallelism
enhancement
New feature or request
#148
opened Jan 28, 2025 by
yzhangcs
[Bug] TypeError: 'constexpr' object is not iterable
bug
Something isn't working
stale
#138
opened Jan 23, 2025 by
York-Cheung
[RFC] Support for more New feature or request
urgent
finetuning Transformers to RNNs
methods (e.g., LOLCATS)
enhancement
#127
opened Jan 19, 2025 by
sustcsonglin
[RFC] Autotune should consider batch size and number of heads
enhancement
New feature or request
urgent
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.