Skip to content

Pull requests: microsoft/onnxruntime

New pull request New
542 Open 15,179 Closed
542 Open 15,179 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
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

add unidirectional attribute to GQA
#23412 opened Jan 17, 2025 by hann-wang • Draft
[MIGraphX EP] Adding Ortvalue features support for MGX EP
#23404 opened Jan 17, 2025 by TedThemistokleous Loading…
9
Optimitze DepthToSpace for mode = DCR, blocksize = 4 and width = 8x
#23399 opened Jan 16, 2025 by yihonglyu Loading…
1
Add trace event control for ORT Web performance profiling ep:WebNN WebNN execution provider
#23393 opened Jan 16, 2025 by qwu16 Loading…
8
[Native WebGPU EP] Add packedQKV and do_rotary attribute support to GroupQueryAttention operator ep:WebGPU ort-web webgpu provider
#23386 opened Jan 16, 2025 by satyajandhyala • Draft
7
Fix einsum operator for empty inputs
#23379 opened Jan 15, 2025 by neNasko1 Loading…
19
[webgpu] Only apply subgroup to intel gen-12lp ep:WebGPU ort-web webgpu provider
#23377 opened Jan 15, 2025 by qjia7 Loading…
2
Ep model merge with external data.
#23374 opened Jan 15, 2025 by BoarQing • Draft
Constant fold initializer for DQ node
#23366 opened Jan 14, 2025 by chilo-ms • Draft
6
[webgpu] Restore MatMulNBits workgroup size for Phi-3.5 ep:WebGPU ort-web webgpu provider
#23349 opened Jan 14, 2025 by daijh Loading…
16
[Test][Don't review] load B twice which means output 16 data in each workgroup
#23334 opened Jan 13, 2025 by qjia7 • Draft
move changes for avx512 blklen32 special layout from liqun/avx-layout
#23331 opened Jan 11, 2025 by liqunfu • Draft
Compile lib instead of executable when checking compiler features
#23329 opened Jan 11, 2025 by bearbones Loading…
2
fixes bug and enables scatternd for jsep ep:WebGPU ort-web webgpu provider
#23311 opened Jan 9, 2025 by prathikr Loading…
1
1
[TensorRT] Support Multiple EP Context
#23294 opened Jan 8, 2025 by jingyanwangms Loading…
19
Upgrade current MacOS-13 to 14
#23293 opened Jan 8, 2025 by jchen351 Loading…
1 task
7
ThreadPool: Spend less time busy waiting. (2nd Attempt)
#23278 opened Jan 7, 2025 by goldsteinn Loading…
[WebNN] Support Cast fusion specific for int64 data type ep:WebNN WebNN execution provider
#23256 opened Jan 6, 2025 by Honry Loading…
10
Model Builder API
#23223 opened Dec 30, 2024 by skottmckay • Draft
46
Add new kernels
#23220 opened Dec 29, 2024 by bili2002 Loading…
21
[js/node] allow arenaExtendStrategy and gpuMemLimit option for CUDA EP
#23176 opened Dec 21, 2024 by nomagick Loading…
@fs-eire
20
Update CUDA install instructions for generate() API
#23155 opened Dec 19, 2024 by natke Loading…
[webgpu] support Pad operator ep:WebGPU ort-web webgpu provider
#23141 opened Dec 18, 2024 by xhcao Loading…
47
Make MultiHeadAttention op return attention probabilities
#23125 opened Dec 16, 2024 by amancini-N Loading…
1
9
[Fix] in Xnnpack EP, the conversion for fused activation param isn't correct
#23115 opened Dec 16, 2024 by mszhanyi Loading…
19
ProTip! Filter pull requests by the default branch with base:main.