Skip to content

Pull requests: flagos-ai/FlagGems

New pull request New
314 Open 1,666 Closed
314 Open 1,666 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

[PerfXLab] optimize fused_moe ops/fused size/Large
#2230 opened Apr 3, 2026 by v2yield Loading…
1
[FlagGems Operator Development Competition] Add grid_sample operator competition This PR is related to gems competition. core ops/aten size/Medium vendor/Ascend
#2229 opened Apr 3, 2026 by katharsis666-max Loading…
6 tasks done
[Bug Fix] Fix_muti_backend_bugs core size/Large tests vendor/AIPU vendor/AMD vendor/ARM vendor/Ascend vendor/Cambricon vendor/Hygon vendor/Iluvatar vendor/Kunlunxin vendor/MetaX vendor/MooreThreads vendor/NVIDIA vendor/Sunrise vendor/TsingMicro
#2226 opened Apr 3, 2026 by Galaxy1458 • Draft
3 tasks
30
Try bring up iluvatar backend test ci/cd core size/Medium size/Small tests vendor/Iluvatar
#2224 opened Apr 3, 2026 by tengqm Loading…
3 tasks
[PerfXLab] Add and optimize gemm op benchmark core ops/aten size/Large tests vendor/NVIDIA
#2220 opened Apr 2, 2026 by bin913 Loading…
3 tasks
@huangyiqun
1
Fix BUG and Optimize performance of mm operator for mthreads backend core size/Medium vendor/MooreThreads
#2219 opened Apr 2, 2026 by Vincent-Xiao Loading…
3 tasks
4
[Cambricon] Optimize 8 ops with larger BLOCK_SIZE kernels size/Medium tests vendor/Cambricon
#2218 opened Apr 2, 2026 by Ankaluoer Loading…
3 tasks
@huangyiqun
[PerfXLab] optimize sqrt op performance size/Small vendor/NVIDIA
#2217 opened Apr 2, 2026 by bin913 Loading…
3 tasks
@huangyiqun
1
[PerfXLab] optimize fill performance benchmark size/Medium vendor/NVIDIA
#2216 opened Apr 2, 2026 by bin913 Loading…
3 tasks
@huangyiqun
1
[SiliconFlow] add Upsample bicubic2d aa backward benchmark core ops/aten size/Large tests
#2215 opened Apr 2, 2026 by ShawnXuan Loading…
3 tasks
[MTHREADS] adapt some config to mt backend size/Small vendor/MooreThreads
#2210 opened Apr 2, 2026 by Oslomayor Loading…
[Advanced Compiler]Add fp8_paged_mqa_logits benchmark core ops/aten size/Large tests
#2208 opened Apr 2, 2026 by AdvancedCompiler Loading…
3 tasks
[KernelGen] Add optimized mul for Iluvatar with 1.16x speedup KernelGen size/Medium vendor/Iluvatar
#2195 opened Mar 31, 2026 by zacliu2023 Loading…
2 of 3 tasks
2
[KernelGen] Add reciprocal operator for Iluvatar platform size/Small vendor/Iluvatar
#2193 opened Mar 31, 2026 by zacliu2023 Loading…
2
[KernelGen] Optimize pow_scalar operator with 3.15x speedup KernelGen size/Medium vendor/Iluvatar
#2188 opened Mar 30, 2026 by zacliu2023 Loading…
1
fix: clamp start/end in slice_backward to prevent out-of-bounds kernel write ops/aten size/Small tests
#2187 opened Mar 30, 2026 by physics31415926 Loading…
@0x45f
2
[KernelGen] Add optimized ones operator for Iluvatar platform KernelGen size/Large vendor/Iluvatar
#2186 opened Mar 30, 2026 by zacliu2023 Loading…
3 tasks done
3
[KernelGen][Ascend] Optimize full op with hand-written multi-core Triton kernel benchmark size/XLarge tests vendor/Ascend
#2185 opened Mar 30, 2026 by Schopenhauer-loves-Hegel Loading…
[KernelGen] Add optimized true_divide operator for Iluvatar platform KernelGen size/Large size/XLarge vendor/Iluvatar
#2183 opened Mar 30, 2026 by zacliu2023 Loading…
8 tasks done
[PerfXLab] optimize mul op ops/aten size/Medium vendor/NVIDIA
#2180 opened Mar 30, 2026 by bin913 Loading…
3 tasks
[KernelGen] Add optimized repeat operator with 1.37x speedup KernelGen size/Large vendor/Iluvatar
#2174 opened Mar 30, 2026 by zacliu2023 Loading…
5 tasks done
[KernelGen] Add view_as_complex operator KernelGen size/Large vendor/Iluvatar
#2169 opened Mar 29, 2026 by zacliu2023 Loading…
4 tasks done
[KernelGen] Add optimized ones_like operator with 1.15x speedup KernelGen size/Large vendor/Iluvatar
#2168 opened Mar 29, 2026 by zacliu2023 Loading…
5 tasks done
[KernelGen] Add optimized add operator with 1.01x speedup KernelGen size/Large vendor/Iluvatar
#2167 opened Mar 29, 2026 by zacliu2023 Loading…
5 tasks done
3
[KernelGen] Add optimized clamp operator with 1.0x speedup KernelGen size/Large vendor/Iluvatar
#2166 opened Mar 29, 2026 by zacliu2023 Loading…
4 tasks done
ProTip! What’s not been updated in a month: updated:<2026-03-03.