Skip to content

Pull requests: ROCm/AMDMIGraphX

New pull request New
87 Open 3,124 Closed
87 Open 3,124 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
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
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
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Added support to set mlir defaults
#4651 opened Mar 4, 2026 by pnikolic-amd Loading…
@pnikolic-amd
Add --replace-literals flag to replace literals with parameters simple small or simple changes
#4649 opened Mar 3, 2026 by pfultz2 Loading…
3
Add flag to strip context simple small or simple changes
#4648 opened Mar 3, 2026 by pfultz2 Loading…
3
Add kernel for fp32 channelwise convolution
#4647 opened Mar 2, 2026 by pfultz2 Loading…
@pfultz2
12
Use JIT Pooling by default and add tuning
#4646 opened Mar 2, 2026 by pfultz2 Loading…
2
[AIMIGRAPHX-581] jit implementation for reverse
#4645 opened Mar 2, 2026 by bdevorem Loading…
@bdevorem
13
Bump sphinx-collapse from 0.1.3 to 0.1.4 in /docs/sphinx documentation
#4644 opened Mar 2, 2026 by dependabot bot Loading…
2
rocMLIR Weekly Sync 2026-03-01 automated Pull request is auto generated by actions dependencies Pull requests that update a dependency file rocMLIR
#4643 opened Mar 1, 2026 by github-actions bot Loading…
@TedThemistokleous
1
fix jit pooling
#4642 opened Feb 27, 2026 by aarushjain29 Loading…
8
Dont use double or int64 when using fast_math
#4641 opened Feb 27, 2026 by pfultz2 Loading…
@kahmed10
9
Onnxruntime Weekly Sync 2026-02-27 automated Pull request is auto generated by actions dependencies Pull requests that update a dependency file onnxruntime PR changes interaction between MIGraphX and Onnxruntime skip bot checks Skips the Performance and Accuracy CI tests
#4640 opened Feb 27, 2026 by github-actions bot Loading…
@TedThemistokleous
yolo26 example
#4639 opened Feb 26, 2026 by alexsu52 • Draft
2
Auto padding convTrans
#4638 opened Feb 26, 2026 by ivarusic-amd Loading…
3
Adding parse for MatMulBnb4 operator and updating MultiHeadAttention operator
#4637 opened Feb 26, 2026 by urpetkov-amd Loading…
@urpetkov-amd
16
Move rotary embedding to op builder
#4634 opened Feb 25, 2026 by pfultz2 Loading…
16
Dont fuse op on attention that are not contained
#4633 opened Feb 25, 2026 by pfultz2 Loading…
4
Use Eigen 3rd party library for ref GEMMs
#4631 opened Feb 24, 2026 by kahmed10 Loading…
11
[AIMIGRAPHX-568] jit implementation for logsoftmax
#4630 opened Feb 24, 2026 by bdevorem Loading…
@bdevorem
16
Add option to skip benchmarking in compile_ops
#4628 opened Feb 24, 2026 by kahmed10 • Draft
4
Add debug symbols for parsed and compiler pass replaced instructions enhancement New feature or request
#4626 opened Feb 20, 2026 by CharlieL7 Loading…
@CharlieL7
45
[AIMIGRAPHX-544] Parallel compilation for dynamic graphs
#4625 opened Feb 20, 2026 by shivadbhavsar • Draft
@shivadbhavsar
Fuse GQA local window as kv-cache attention
#4617 opened Feb 17, 2026 by turneram Loading…
8
[AIMIGRAPHX-544] Parallel compilation for dynamic graphs
#4616 opened Feb 17, 2026 by shivadbhavsar • Draft
@shivadbhavsar
6
[7.2.1] Backport Inference Server improvements
#4613 opened Feb 13, 2026 by causten Loading…
4
Use rocBLAS GEMV for skinny GEMM (M=1 or N=1) to improve performance
#4608 opened Feb 12, 2026 by klin2024 • Draft
6
ProTip! Adding no:label will show everything without a label.