Skip to content

Issues: pytorch/ao

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

112 Open 73 Closed
112 Open 73 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Unify scaled INT8 matmul
#862 opened Sep 10, 2024 by gau-nernst
Add MyPy support enhancement New feature or request
#856 opened Sep 9, 2024 by drisspg
Create report for Autoquant autoquant
#855 opened Sep 9, 2024 by drisspg
Add Autoquant Cache autoquant
#854 opened Sep 9, 2024 by drisspg
1
Update Eval scripts enhancement New feature or request
#853 opened Sep 9, 2024 by drisspg
int4_weight_only dequant does not restore original shape
#848 opened Sep 9, 2024 by jerryzh168
AO dtype composability tracker
#844 opened Sep 8, 2024 by msaroufim
2 tasks
1
[Quantization FSDP] Support quantize_() for DTensor
#803 opened Sep 4, 2024 by gau-nernst
1
NotImplementedError: aten.linear.default not implemented when using MXTensor mx
#796 opened Sep 3, 2024 by Ali-Flt
1
7
placeholder for float8 README.md images
#793 opened Sep 2, 2024 by vkuzo
1
fpx test failing on main
#792 opened Sep 2, 2024 by msaroufim
@jerryzh168
3
'FakeTensor' object has no attribute 'layout_type'
#789 opened Sep 2, 2024 by WAS-PlaiLabs
4
[Question] Is there any plans to support fp8 batched matmul (_scaled_bmm) float8
#781 opened Aug 30, 2024 by leeeizhang
How does this work with ONNX export and quantization? export inference
#777 opened Aug 29, 2024 by ogencoglu
1
All code snippets on README should be self contained or link to single file runnable code
#765 opened Aug 27, 2024 by msaroufim
2 tasks
quantize_(m, int4_weight_only()) is a footgun
#764 opened Aug 27, 2024 by msaroufim
@HDCharles
How to reduce autoquant compilation time
#763 opened Aug 27, 2024 by msaroufim
@HDCharles
1
Shape mismatch error for int8_weight_only quantization when input is of shape (N, 1) bug Something isn't working inference
#761 opened Aug 27, 2024 by jbschlosser
@jerryzh168
First-time user feedback on quantize_ and autoquant
#758 opened Aug 27, 2024 by vkuzo
@msaroufim
2
Slow compile times tracker
#754 opened Aug 27, 2024 by msaroufim
3 tasks
[Feature Request] Fused fp8 matmul kernel (quant dequant matmul)
#752 opened Aug 27, 2024 by qingquansong
1
Outdated gpt-fast codebase autoquant bug Something isn't working
#751 opened Aug 27, 2024 by YJYJLee
2
MoE example good first issue Good for newcomers
#729 opened Aug 22, 2024 by msaroufim
1
What is the difference between WeightNormSparsifier and torch.nn.utils.prune.l1_unstructured ? question Further information is requested
#724 opened Aug 21, 2024 by mayank64ce
@jcaip
3
Two versions of model eval code?
#718 opened Aug 20, 2024 by jerryzh168
ProTip! What’s not been updated in a month: updated:<2024-08-11.