Skip to content

Issues: iree-org/iree-turbine

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

81 Open 31 Closed
81 Open 31 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

[python] Using python version incompatible iree-base-compiler or iree-base-runtime results in obscure errors
#820 opened May 9, 2025 by dan-garvey
MoE Tasklist
#816 opened May 9, 2025 by harsh-nod
6 tasks
examples/aot_mlp/mlp_export_simple.py raises Fatal Python error
#804 opened May 7, 2025 by gilron31
1
Buffer Ops Reduce Performance on Paged Decode Attention
#801 opened May 6, 2025 by harsh-nod
Wave Documentation
#797 opened May 6, 2025 by raikonenfnu
14 tasks
Wave June 2025 Release
#792 opened May 5, 2025 by harsh-nod 100+
Precondition add_get_results of expansion pass is not checked
#791 opened May 5, 2025 by martin-luecke
Core kernels to implement for Wave
#790 opened May 4, 2025 by raikonenfnu
6 tasks
Calling 2 consective get_kernels and then launching the kernels fails
#770 opened Apr 29, 2025 by harsh-nod
IREE compilation flag IREE_ENABLE_RUNTIME_TRACING is breaking import iree.turbine.aot
#746 opened Apr 23, 2025 by voonter
5
aot.export fails for complex-valued literals
#744 opened Apr 23, 2025 by MaxS1996
Flaky CI Tests
#736 opened Apr 22, 2025 by saienduri
1
Speculative Sampling
#715 opened Apr 15, 2025 by raikonenfnu
10 of 38 tasks
2
Summer 2025 Intern Projects
#713 opened Apr 15, 2025 by harsh-nod
21 tasks
[BOO] boo_conv gradcheck test failing with bias add
#704 opened Apr 11, 2025 by zjgarvey
@zjgarvey
1
[export] Exporting MoE block leads to CONSTRAINT_VIOLATION error
#702 opened Apr 11, 2025 by archana-ramalingam
1
Segmentation fault when exporting BatchNorm with aot
#693 opened Apr 8, 2025 by Hacker1337
test case failure with "AssertionError: Dynamic dims not currently supported"
#672 opened Apr 2, 2025 by FlintWangacc
Wave May 2025 Issue
#667 opened Apr 1, 2025 by harsh-nod 100+
fail to compile deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B model
#659 opened Mar 30, 2025 by FlintWangacc
Implement same thing as wave.template.layers.attention.scaled_dot_product_attention but for fp8 quantized variant.
#657 opened Mar 29, 2025 by raikonenfnu
Implement a reference wave.template.layers.attention.scaled_dot_product_attention function that smartly handle multiple batch dims (i.e BSD, BHSD, B_1 B H S D, etc), handling of dynamic dimensions, and using LRU cache inside to prevent recompile or even hashing all together.
#656 opened Mar 29, 2025 by raikonenfnu
Refactor quantized attention kernel to have dynamic seqlen to prevent recompile for LLM
#655 opened Mar 29, 2025 by raikonenfnu
Add support for GEMM fused with bias (currently stalled/broken because for single output reduction we propagate out the index of the whole reductionOp which consist of all the dimensions inside it as opposed to just the index layout of the yielded value mma in this case.)
#654 opened Mar 29, 2025 by raikonenfnu
@xintin
QOP per tensor scaling FP8 GEMM kernel (important for quantized linear layer)
#653 opened Mar 29, 2025 by raikonenfnu
ProTip! Updated in the last three days: updated:>2025-05-09.