Skip to content

Pull requests: buddy-compiler/buddy-mlir

New pull request New
108 Open 447 Closed
108 Open 447 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

Dev/deepseek dynamic input sequence length
#638 opened Dec 3, 2025 by ArCyanic Loading…
[Frontend] Add Comprehensive Core Aten IR Operator Support
#637 opened Dec 3, 2025 by sdh1014 Loading…
[IME] Add SpacemiT IME dialect with LLVM lowering and intrinsic support.
#636 opened Dec 2, 2025 by wujing215 Loading…
10
[buddy-examples] WIP: fix a series of bugs and add features to support various E2E deployments on FPGA.
#635 opened Nov 29, 2025 by shirohasuki Loading…
[Frontend] Apply FlashAttention for DeepSeekR1 Prefill Phase
#634 opened Nov 28, 2025 by GuoningHuang Loading…
1
[Midend] Add INT8 support to AMX matmul pass with unified dtype detection
#633 opened Nov 28, 2025 by Jiajun-Ji Loading…
Add LlamaMode matmul for decode
#632 opened Nov 27, 2025 by Yssx-g Loading…
[MLIR-JIT] Create a patch to fix MLIR-JIT for RISC-V
#628 opened Nov 21, 2025 by ArCyanic Loading…
[Example] Integrate AMX BF16 matmul pass into DeepSeek-R1 inference
#627 opened Nov 21, 2025 by Jiajun-Ji Loading…
[frontend] add new Operator fusion
#626 opened Nov 19, 2025 by Old-cpu Loading…
Fix: Resolve BF16 inference issues for DeepSeek-R1 and introduce PD separation.
#625 opened Nov 19, 2025 by Jiajun-Ji Loading…
Evaluate Transpose and optimize batch_matmul_transpose_b
#616 opened Nov 9, 2025 by xvyv99 • Draft
[examples] Add performance testing for linalg.matmul operator using three passes
#614 opened Nov 9, 2025 by sdh1014 Loading…
2
【frontend】Add graph splitting infrastructure compatible with DeepSeek model
#610 opened Nov 6, 2025 by asdf1113 Loading…
[examples] Add MLIR OMP attributes to control thread count and NUMA affinity
#602 opened Oct 29, 2025 by sdh1014 Loading…
[Frontend] Achieve norm optimization by scf dialect and intergrate into E2E pass
#596 opened Oct 27, 2025 by GuoningHuang Loading…
1
[midend] Add FuseFC pass
#589 opened Oct 21, 2025 by ArCyanic Loading…
[Frontend] Implement Online LayerNorm with Frontend Fusion and Lowering Pass Support
#587 opened Oct 20, 2025 by GuoningHuang Loading…
8
Add a pass that fuses matmul and transpose operations
#567 opened Oct 2, 2025 by FloatingcloudKnight Loading…
11
feat: Add BLIS-like GEMM implementations
#565 opened Sep 29, 2025 by ShiHaoGao Loading…
[subsection_tag]tosa test by linzhen
#563 opened Sep 27, 2025 by linzhenpl07 Loading…
3
[Midend] Add softmax optimize midend pass.
#560 opened Sep 27, 2025 by Jiajun-Ji • Draft
[tests] Add VIR Python binding and simple tests for it
#555 opened Sep 24, 2025 by ArCyanic Loading…
[Frontend] Lower attention op to FlashAttention vectorized kernel. enhancement New feature or request format issue
#553 opened Sep 24, 2025 by GuoningHuang Loading…
5
Add RVV cross-compilation support for DeepSeekR1 Example
#549 opened Sep 18, 2025 by R-Tars Loading…
ProTip! Type g i on any issue or pull request to go back to the issue listing page.