Skip to content

Issues: ROCm/aiter

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

16 Open 26 Closed
16 Open 26 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

[Issue]: aiter.fused_moe_bf16_asm.ck_moe_2stages is not behaving like aiter.asm_moe in FP8 per tensor dynamic quantization
#338 opened Apr 24, 2025 by tjtanaa
1
[Question] Does aiter support FA3?
#330 opened Apr 23, 2025 by danielhua23
[Feature]: gfx1201 support
#321 opened Apr 19, 2025 by l33tm4st3r
[Feature]: fmoe tkw1 suppprt for bf16
#298 opened Apr 15, 2025 by sijiac
@lalala-sh
test_mha.py:test_mha_backward fails with causal=True and seqlen_q < seqlen_k bug Something isn't working triton
#289 opened Apr 14, 2025 by juuso-oskari
@juuso-oskari
[Feature]: enable fuse topk_weight at moe_stage1
#286 opened Apr 14, 2025 by valarLip
2 tasks
@junhaha666 @lalala-sh
[Feature]: Support torch compile graph mode for custom ops - trigged by vLLM V1 and aiter
#283 opened Apr 10, 2025 by hongxiayang
1
[Issue]: CK attention still doesn't work on 7900 XTX?
#254 opened Mar 30, 2025 by FeepingCreature
7
[Documentation]: Fill out supported operators in README
#246 opened Mar 26, 2025 by garrettbyrd
@Luciacao
[Issue]: Some AITER Operators are not compatible with torch.compile
#244 opened Mar 25, 2025 by tjtanaa
@junhaha666
7
[Issue]: test_moe_2stage produces wrong results
#212 opened Mar 18, 2025 by cognaiger9
1
[Help] Does branch triton_mha_fwd_kernel should suport rdna2/gfx1030 amd gpu?
#210 opened Mar 17, 2025 by inevity
2
[Feature]: support seqlen_q large than 1 for PagedAttention with FP8KV
#181 opened Mar 6, 2025 by sijiac
[Issue]: Build errors for gfx90a (MI250) architecture
#179 opened Mar 5, 2025 by OscarSavNS
4
Inference attention tech path
#79 opened Jan 27, 2025 by carlushuang
1
deepseek-v3/R1 tech path
#75 opened Jan 23, 2025 by carlushuang
1
ProTip! Follow long discussions with comments:>50.