Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
549 commits
Select commit Hold shift + click to select a range
e945b03
[GPU] Fix FP16 overflow in fsv16 MVN kernel (#35942)
sungeunk May 19, 2026
d7c4dcf
[gpu][paged_attention]: Optimized bidirectional mask range calculatio…
pkowalc1 May 19, 2026
57731f4
[GPU]fix error handler (#33810)
openvino-dev-samples May 19, 2026
88f5a5d
[CI][GHA][RV64] Switch from Xuantie to riscv-collab RISC-V toolchain …
aobolensk May 19, 2026
fe713f1
[NPUW][Tests] Add skip posibility for `SerializationTestNPUW` (#35978)
StefaniaHergane May 19, 2026
cf5ce82
[GPU] Robust INT8 dynamic quantization handling for hybrid linear-att…
andrew-k-park May 19, 2026
e923f6b
[CI] [GHA] Add CI Doctor for Merge Queue (#35979)
akashchi May 19, 2026
abc3de3
[ONNX] Store decoder with shared_ptr instead of raw pointer (#35939)
mvafin May 19, 2026
eae8f58
[GPU] Remove unused variable from CM PA generator (#35984)
maxnick May 19, 2026
f62a7a9
[CI] [GHA] Use Ubuntu 22 for py3.13 tests in manylinux Python API job…
akashchi May 19, 2026
28bc5d7
[NPU] Don't try to destroy the graph twice (#35990)
pereanub May 20, 2026
28217a1
[CPU] enable adaptive r-kv algorithm support (#35254)
azhai219 May 20, 2026
689f06c
[GPU] fix dGPU func testcases in smoke_ScaledAttnDynamic4D_GPU (#35437)
yuanxion May 20, 2026
a2881a0
[GHA] Fix pip installation for Python 3.9 in ARM64 Ubuntu 22 images (…
ababushk May 20, 2026
37451e0
[Transformations] Enable FuseGroupedQueryIntoGDN for statically-shape…
mryzhov May 20, 2026
31ea44c
[GPU] Add Sign activation support for integer types (#36003)
jade-cho May 20, 2026
b1dd40f
[NPUW] Enabling GQA for fused hfa (#35741)
kkoryun May 20, 2026
b9bad1e
[GPU] Fix lora/moe kernel compilation on modern NEO drivers (Gen 9.5+…
pjordanandrsn May 20, 2026
1bbe690
[TRANSFORMATIONS] Extend Sliding Window pattern for Gemma3-270m with …
CuriousPanCake May 20, 2026
6a1298f
[GPU] validate_output: fp16 input range check for sin/cos (#35983)
ahnyoung-paul May 21, 2026
b450859
Bump eslint from 10.3.0 to 10.4.0 in /samples/js/node in the npm-pack…
dependabot[bot] May 21, 2026
302b14e
[Level Zero] fixes for Zero Api dependencies (#36011)
Roszczyk May 21, 2026
80736f9
[TEMPLATE] Fix GatherND evaluate on TEMPLATE plugin (#35988)
praasz May 21, 2026
91e85d3
[Snippets][CPU][RV64] Implement binary call emitter and perf count em…
aobolensk May 21, 2026
e9e47c2
Bump certifi from 2026.4.22 to 2026.5.20 (#36025)
dependabot[bot] May 21, 2026
f7f2287
[ONNX FE] Wrap SequenceInsert output in SequenceMark (#36015)
mvafin May 21, 2026
347e5a8
[core] Remove OwningSharedStreamBuffer class (#35845)
praasz May 21, 2026
f81f692
[GPU] Align GatherMatmul behavior with moe_gemm (#35780)
EgorDuplensky May 21, 2026
edc9f5d
CVS-175243 use secondary hf token openvino (#35701)
jkoniusz May 21, 2026
d276f51
[`NPU`] [`ENCRYPTION`] Throw OV exception instead of warning for olde…
MirceaDan99 May 21, 2026
ee466aa
[Snippets][CPU][RV64] Preserve CPU runtime configurator in CPUTargetM…
aobolensk May 21, 2026
a49eda1
Bump brace-expansion from 5.0.5 to 5.0.6 in /samples/js/node (#35973)
dependabot[bot] May 21, 2026
1467b0f
[Snippets][CPU][RV64] Add Select emitter (#35957)
aobolensk May 21, 2026
b1aec35
[CPU][Snippets] Fix clang-tidy with disabled debug caps (#35559)
v-Golubev May 21, 2026
0b01f28
[Snippets][CPU][RV64] Skip loop begin emission for static evaluate_on…
aobolensk May 21, 2026
27f582b
[TRANSFORMATIONS] Remove marking KV-cache parameters with enable_keep…
CuriousPanCake May 21, 2026
9492e77
[GPU] Fix to clone shared Convert always in ConvertMaMulToFullyConnec…
davidsnam-intel May 22, 2026
ad63ea4
[GPU] doc update for L0 test (#36034)
isanghao May 22, 2026
7bea460
[PyOV] Add check for returned empty map from get_property (#36009)
alexandruenache1111 May 22, 2026
d039da4
[GPU] fix CL_OUT_OF_RESOURCE issue in arg_max_gpu_min_large_output_si…
yuanxion May 22, 2026
8da2cf4
[NPU] Remove support for exclusive_async_requests property from NPU (…
pereanub May 22, 2026
d802d32
Add RMSNormalization & RotaryEmbedding (opset 23) and shared decompos…
mvafin May 22, 2026
b6966a3
[CPU] Support PagedAttentionExtension with Shared KV-cache (#35561)
v-Golubev May 22, 2026
a9db3ae
[LPT] Remove mandatory weight folding in ConvolutionTransformation (#…
mdvoretc-intel May 22, 2026
2986504
[GPU] Enable SDPA Fusion for 3D inputs and implicit GQA broadcasting …
clee30 May 23, 2026
935d654
[TRANSFORMATIONS] Add EliminateConvPaddingMaskGating pass to SDPAToPA…
CuriousPanCake May 25, 2026
7f129fa
[core] Implement aligned memory allocation utilities and tests (#35872)
praasz May 25, 2026
936aa06
Choose dynamic graph backend based on blob header (#35996)
ArtemySkrebkov May 25, 2026
e1aeef5
[TRANSFORMATIONS] Fix EliminateIdentityConvert behavior with multiple…
CuriousPanCake May 25, 2026
e019397
[GPU] Optimize performance for moe with speculative decoding (#35901)
riverlijunjie May 26, 2026
dfce999
[NPUW] Covered Linear Attention support with tests (#35748)
AsyaPronina May 26, 2026
86048d9
[PA ref] Add support for bidirectional image attention (#36027)
p-wysocki May 26, 2026
fe77702
[NPUW] Split model builder into per-component files (#35735)
dylanneve1 May 26, 2026
9bb2332
[CPU] Fix/rmsnorm avx512 zero trip (#35912)
tiger100256-hu May 26, 2026
5f343ea
[NPUW] Fix unit test compilation issue (#35482)
DingZhangIntel May 26, 2026
f9766fb
[GPU] Enable default 4bit kv cache (#35703)
byungilm May 26, 2026
8150e9f
[inference] Identify failing port in check_tensor shape mismatch erro…
evkotov May 26, 2026
2b6aa4d
Fix AArch64 multi-ISA LTO codegen isolation (#35994)
allnes May 26, 2026
304d13b
Bump soupsieve from 2.8.3 to 2.8.4 (#36055)
dependabot[bot] May 26, 2026
0b0029e
[CPU] Refactor GatherMatmul to implement executor design (#35812)
maxnick May 26, 2026
8a743bb
[NPU] Deprecate ENABLE_CPU_PINNING on NPU and remove logic for it (#3…
pereanub May 26, 2026
6c253ae
[HETERO] Fix cycle-splitting failure when shared constants bridge sub…
wgzintel May 27, 2026
eef2a81
feat: Add Copilot agent instructions (migrated from OMEGA) (#35322)
mlukasze May 27, 2026
512e925
[TRANSFORMATIONS] Fix shared constant reshape in MoveEltwiseUpThrough…
zhanmyz May 27, 2026
e5561b0
[gha] Fix and increase timeout for Linux RISC-V (#36089)
ababushk May 27, 2026
9a3a705
[Transformations] Shared low_precision_dequantize decomposition helpe…
mvafin May 27, 2026
bf0228a
[CPU] Enable BF16 dynamic quantization path for compressed FullyConne…
liubo-intel May 27, 2026
d84ebb9
[CPU][ARM] Implement FuseClampAndFakeQuantize transformation (#36042)
alvoron May 27, 2026
bd22d2b
[Snippets][CPU][ARM] Use typed GEMM executor dispatch (#35747)
aobolensk May 27, 2026
51b2fef
[GPU] support input/output padding for reorder_data_fsv (#36037)
sungeunk May 27, 2026
3c9dfab
[NPU] Create a separate compiler per compiled model (#36091)
pereanub May 28, 2026
c93f27e
[Protopipe] Update opencv to fix input tensor layout parsing for U8 m…
svigh May 28, 2026
1e4690c
[GPU] Fix Kokoro TTS FP16 silent / NaN audio output (#36064)
andrew-k-park May 28, 2026
44e2867
[`NPU`] [`TESTS`] [`ENCRYPTION`] Fix internal encryption tests throwi…
MirceaDan99 May 28, 2026
342e048
Python: Expose GQA Extension in Python bindings (#36054)
dmatveev May 28, 2026
f87d5d4
[CI] [GHA] Remove gstreamer from Debian ARM Dockerfile (#36107)
akashchi May 28, 2026
8ed0432
[NPUW] Fix longrope for Phi models (#35981)
eshiryae May 28, 2026
cb5f5cc
[GPU] Fallback unaligned blocked concat to OCL (#36079)
sungeunk May 28, 2026
5ff2fbd
[GHA] Enable usage of Docker images built in ACR on HRZ Linux runners…
gkoscins May 28, 2026
dcac25c
Remove redundant jit declaration (#36083)
byungilm May 28, 2026
97344a7
NPUW: Introduce a new FOLD_ONLY options (#36060)
dmatveev May 28, 2026
ac001ba
[TRANSFORMATION] Introduce new DisablePrecisionConversion RT attribut…
CuriousPanCake May 29, 2026
d790238
Pin `kernels` Python module version to fix PyTorch Models Tests (#36136)
ababushk May 29, 2026
c24dea3
[Snippets][CPU][ARM] Use KAI RHS layout for GemmCopyB loop ports (#35…
aobolensk May 29, 2026
6c6f45f
[TRANSFORMATIONS] Refactor/Improve MoveEltwiseUpThroughDataMovScalar …
CuriousPanCake May 29, 2026
a20bb26
[Build] Fix gcc15 build issues for OpenVINO build (#36030)
praasz May 30, 2026
44f104f
[GPU] dump_memory_pool improvement (#36130)
isanghao May 31, 2026
23acad8
CVS-167265-Support dilation>1 for ConvolutionBackpropData and GroupCo…
CailiWuIntel Jun 1, 2026
53582cf
[CPU] Tree attention (#35352)
songbell Jun 1, 2026
3cf8227
[GPU] modify matching patterns for increase_position_ids_for_ltx_vide…
ahnyoung-paul Jun 1, 2026
2716a0f
[PyOV] Update Python API stub files (#36155)
sys-openvino-ci Jun 1, 2026
d202a93
[CPU][ARM] Upgrade ACL to 53.1.0 (#36096)
alvoron Jun 1, 2026
3224593
Bump openvino-node from 2026.1.0 to 2026.2.0 in /samples/js/node in t…
dependabot[bot] Jun 1, 2026
3fe8ef1
[GPU] Add SYCL runtime support (#35725)
tknose Jun 1, 2026
405ed25
disable eltwise fusion to FC if connected to input_hidden_states (#35…
nazanin-beheshti Jun 1, 2026
5283c55
[GPU] Add dxgi mem tracker (#35821)
hyunback Jun 1, 2026
5380884
[GPU] Fix testcases of write lock_type issue on B580 (#35493)
yuanxion Jun 1, 2026
0a65b43
[GHA] [CI] precommit dgpu test for L0 ubuntu24 (#36158)
gkoscins Jun 1, 2026
e417633
[GHA] [CI] Enable L0 Windows build (#36159)
gkoscins Jun 1, 2026
16f03ba
[GHA] CC workflow simplified for Linux (#36163)
yury-intel Jun 1, 2026
0bee870
[CPU][ARM] ACL-specific int8 AvgPool cases handling (#36117)
alvoron Jun 1, 2026
a6efe11
Fix XAttention for mixed multi-sequence prefill/decode test cases (#3…
WeldonWangwang Jun 1, 2026
7924859
[GPU] int4-kv-cache for blog posting (#36148)
isanghao Jun 2, 2026
ddaa88e
[TEST] Activate save-report retry timeout for functional tests (#35869)
zhanmyz Jun 2, 2026
4d306f3
[GPU] Fix weight reorder index calculation for grouped convolutions w…
maxnick Jun 2, 2026
8c8c67c
[Snippets][CPU][RV64] Enable Ceiling emitter (#36149)
aobolensk Jun 2, 2026
b24a1a7
Bump the npm-packages group across 1 directory with 3 updates (#36123)
dependabot[bot] Jun 2, 2026
7acedf1
[NPU] New vcl_allocator_3 to support multiple allocations and robust …
ShaojieZhuIntel Jun 2, 2026
b971545
[SDPAToPA] Clean up `token_type_ids` detection (#35938)
p-wysocki Jun 2, 2026
05faed4
[GHA][Coverage] coverage_toolkit composite action (#35688)
mkruszyx Jun 2, 2026
7fe2854
[NPUW] Introduce KVCacheBlockManager for block-based KV cache. (#35013)
intelgaoxiong Jun 2, 2026
c31cd6b
[GPU] Introduce a parallelized NonZero implementation (#35239)
mdvoretc-intel Jun 2, 2026
b7a6941
[NPUW]Fix MoE downstream subgraph dump. (#36129)
intelgaoxiong Jun 2, 2026
7c21039
Bump eslint from 10.4.0 to 10.4.1 in /samples/js/node in the npm-pack…
dependabot[bot] Jun 2, 2026
16207e3
[GPU][Level Zero] Names unification L0->ZE (#36067)
Roszczyk Jun 2, 2026
ffae04b
[GPU] Fix incorrect INT8 convolution results caused by _sub_group_shu…
yuanxion Jun 2, 2026
bd98149
[core] Create lazy loading buffer (#35512)
t-jankowski Jun 2, 2026
5f62a55
[Transformations] Fix ConvertFCToCompressed apply_transpose for rank-…
mvafin Jun 2, 2026
8267f8d
Add Intel Copyright to Dockerfiles (#36097)
ValentinaKats Jun 2, 2026
459e9ac
Disable DDA and Auto Unrolling for InferWithHostCompileTests tests (#…
MikhailPedus Jun 2, 2026
66db6ac
[ONNX FE] Support allowzero in Reshape (#36046)
mvafin Jun 2, 2026
23fd5d9
[OV JS] Fix GC safety for zero-copy tensor construction from TypedArr…
Retribution98 Jun 2, 2026
2681144
[GPU] Fix CL_OUT_OF_RESOURCES for 1D convolution with large filter on…
jade-cho Jun 3, 2026
5ecea94
[GPU] Add driver-independent regression test for __local block-read k…
pjordanandrsn Jun 3, 2026
eac4b44
[NPU] Create NPU executor class (#36013)
pereanub Jun 3, 2026
dec3dd1
[SIT] Add prefill options for nrmse (#36188)
svigh Jun 3, 2026
a959d01
[NPUW]Fixed high memory footprint with blob cache. (#36195)
intelgaoxiong Jun 3, 2026
a8892bd
[NPU] Enable tests after NPU driver updated (#36043)
StefaniaHergane Jun 3, 2026
bdf4a17
[CI][GHA] Enable treating warnings as errors for Linux RISC-V job (#3…
aobolensk Jun 3, 2026
12a2726
[core] Add constant data evict on Windows with mmap (#35570)
praasz Jun 3, 2026
7ba34b7
Support f32 input and bf16 output in Gather JIT kernel (#36004)
mangguo321 Jun 3, 2026
5e24a79
[GPU][TESTS] Removed debug prints (#36208)
v-Golubev Jun 3, 2026
918b836
[CPU][ARM] Call `ConvolutionTransformation` transformation for "Conv-…
alvoron Jun 3, 2026
c68e629
[LevelZero] Bump up level zero repository to v1.29.0 (#36206)
pereanub Jun 3, 2026
9e6d8a9
[GPU] delete _SECURE_SCL (#36048)
michal-miotk Jun 3, 2026
d0940ab
[GPU] Fix implicit concat fusing blocked for feature-axis and multi-u…
deepaks2 Jun 3, 2026
3a3ac6c
[ONNX FE] Add com.microsoft.BifurcationDetector translator (#36047)
mvafin Jun 3, 2026
4c59e69
[Transformations] Added a new `TransposeFQ` matcher to common transpo…
alvoron Jun 3, 2026
f144078
[GPU][CM] Fix unused lambda capture in paged_attention_gen (#36185)
sungeunk Jun 4, 2026
20d0d8e
[GPU] Fix null node guard in GPTOSS precision pass (#36245)
sungeunk Jun 4, 2026
0b2e4e1
[GPU] Fix perf regression in fc_bf_tiled DQ scale path by using f16 s…
sungeunk Jun 4, 2026
367f60d
[OV MX][GPU] Fix MoE model cache load crashes (#36152)
xzhan34 Jun 4, 2026
265cb97
[NPUW] Fix GQA broadcast shape to match AttentionBroadcast pattern (#…
dylanneve1 Jun 4, 2026
ce500e9
[NPUW][MoE] Fix missing MoE runtime behavior attachment on cache impo…
intelgaoxiong Jun 4, 2026
47ee139
[NPUW] Fix PreserveConstDictMatMulAsymm to match asymmetric quantized…
xu-shaobo Jun 5, 2026
c86befc
[GPU] Enable mmap for loading model weights​ on iGPUs (#34494)
Prithviraj-R Jun 5, 2026
791fa94
[GPU] Fix dynamic quantization large group case (#36147)
chenhu-wang Jun 5, 2026
9e4ea4f
[GPU] Improve gdn kernel via vload & fma (#36248)
zhangYiIntel Jun 5, 2026
0b706cf
[CI] [GHA] Increase CI Doctor MQ memory retention period (#36262)
akashchi Jun 5, 2026
7932a11
Bump typescript-eslint from 8.60.0 to 8.60.1 in /src/bindings/js/node…
dependabot[bot] Jun 5, 2026
8765d11
[Snippets][CPU][RV64] Enable dynamic Power emitter (#36180)
aobolensk Jun 5, 2026
6e52fdd
[Snippets][CPU][RV64] Optimize jit_fill_emitter (#36259)
aobolensk Jun 5, 2026
94bad43
[GPU] Fix gemm_tiled_opt incorrect results for 1D transposed weight (…
jade-cho Jun 8, 2026
53e932a
[Tests] Enable FP16 conversion in OVC (#36266)
dorloff Jun 8, 2026
c2262d2
[GPU] Add ReduceFCDimensions transformation (#36212)
mdvoretc-intel Jun 8, 2026
206e79f
[Snippets][CPU][RV64] Implement load/store + convert fusion (#36156)
aobolensk Jun 8, 2026
a7ba8eb
[NPU] Disable shared common queue feature (#36264)
pereanub Jun 8, 2026
4bb3a8f
[GPU] Fuse extra transpose from compressed conv1x1 transformation (#3…
mdvoretc-intel Jun 8, 2026
340a9aa
[TF FE] Indirect leak on ONNX FE - variables_index.cpp (#36087)
bumbosiepsak Jun 8, 2026
e3c7809
Support torch_fused GPTQ quant type (#36132)
evkotov Jun 8, 2026
6b75049
[GHA] Enable func tests for L0 ubuntu24 build (#36291)
gkoscins Jun 8, 2026
2cafa5d
[Snippets][CPU][RV64] Optimize jit_horizon_emitter (#36258)
aobolensk Jun 8, 2026
55076c4
[Transformations] Relax conversion to fp16 relative error (#36005)
mryzhov Jun 8, 2026
c48e09e
[NPU] Disable test before NPU driver updated (#36308)
StefaniaHergane Jun 8, 2026
fb0fef7
[PDPD] fix Fuzzing negative-size-param (#36256)
azhai219 Jun 9, 2026
ed4cfb1
Fix xattention find_block (#35697)
WeldonWangwang Jun 9, 2026
51ae979
[GPU] Fix null dereference risk in GPU cache load zero-copy path (#36…
sungeunk Jun 9, 2026
3654766
[NPUW][MoE Scaling]Support MoE expert layout variants (#36184)
intelgaoxiong Jun 9, 2026
d978369
[ONNX FE][Transformation] GroupQueryAttention refine work for static …
bopeng1234 Jun 9, 2026
4462a3b
[NPU] Remove npu_enable_strides_for from supported properties if devi…
pereanub Jun 9, 2026
dfe9b38
[GPU] Fixing convolution fsv16 kernels to avoid crashes in Level Zero…
Roszczyk Jun 9, 2026
a26fe9f
[NPUW] Remove DownsampleInterpolate avoid pattern as unused (#36278)
OrestChura Jun 9, 2026
64cdd25
[CI] [GHA] Use smaller model for CI Doctor (#36325)
akashchi Jun 9, 2026
eed90e7
[FE] Add Sequence{At,Erase,Length} framework helpers (#36228)
mvafin Jun 9, 2026
94dd7eb
Bump the npm-packages group across 1 directory with 2 updates (#36317)
dependabot[bot] Jun 10, 2026
e702c64
Bump idna from 3.13 to 3.15 in /docs (#35995)
dependabot[bot] Jun 10, 2026
6527099
Bump snowballstemmer from 3.0.1 to 3.1.1 (#36242)
dependabot[bot] Jun 10, 2026
18ebd09
[CI] [GHA] Use `repo-memory` to store CI Doctor MQ investigations (#3…
akashchi Jun 10, 2026
ec5d4c7
[Snippets][CPU][RV64] Enable JIT EltwiseSelect execution using existi…
aobolensk Jun 10, 2026
eb6ba43
Bump mistune from 3.2.0 to 3.2.1 (#35657)
dependabot[bot] Jun 10, 2026
a5939eb
Bump zipp from 3.23.1 to 4.1.0 (#36000)
dependabot[bot] Jun 10, 2026
9246c6a
[NPU] UD2026.28 Engineering Drop ww23 (#36225)
Maxim-Doronin Jun 10, 2026
514fc28
[PT FE] Upgrade tests to torch 2.12 (#36098)
mvafin Jun 10, 2026
bdf6935
[ONNX FE] Support ScatterND reduction attribute (#36072)
mvafin Jun 10, 2026
de4e7dd
[CPU][Tests] Fix params for Power eltwise test inputs generator (#36332)
aobolensk Jun 10, 2026
23b173e
[CPU][Tests] Remove duplicate eltwise test declarations and suites (#…
aobolensk Jun 10, 2026
3c2e616
[GPU] WA: Fix attention sink handling in FlashAttnV2 PA SDPA path (#3…
wilson-seok Jun 11, 2026
428e583
[GPU] Fix OOB reads in convolution bfyx_os_iyx_osv16 for multi-batch …
davidsnam-intel Jun 11, 2026
337f637
[test] Add read_model onnx fuzzer (#34433)
jszczepa Jun 11, 2026
bd296f5
[CI] [GHA] Compile CI Doctor MQ agentic workflow (#36351)
akashchi Jun 11, 2026
b9a44e4
Bump cython from 3.2.4 to 3.2.5 (#36057)
dependabot[bot] Jun 11, 2026
eb884f5
[GPU] tensor from nthandle (#35543)
michal-miotk Jun 11, 2026
a86a56d
[NPU] Mark cancel() method for inferences unsupported (#36340)
pereanub Jun 11, 2026
423bd28
[GPU] MoE 3 GeMM: separate Router Subgraph and MoE body kernels (#35691)
v-Golubev Jun 11, 2026
b96765b
[HETERO] Fix multi-hop cycle in SubgraphCollector (#36126)
wgzintel Jun 11, 2026
0d9fa94
[GPU] check nullptr in add_output_chain (#36347)
michal-miotk Jun 11, 2026
2b5c69a
[GPU] Fix for ExperimentalDetectronROI levels (#36305)
nshchego Jun 11, 2026
ba9ed7a
[Transformations] Enhance unsqueeze axis detection in reshape transfo…
praasz Jun 11, 2026
fc07431
Bump actions/stale from 10.2.0 to 10.3.0 (#36022)
dependabot[bot] Jun 11, 2026
384e5f9
Bump idna from 3.13 to 3.18 (#36243)
dependabot[bot] Jun 11, 2026
d22abf8
Bump requests from 2.33.1 to 2.34.2 (#35911)
dependabot[bot] Jun 11, 2026
0117a24
Bump urllib3 from 2.6.3 to 2.7.0 in /docs (#35791)
dependabot[bot] Jun 11, 2026
4089686
[CORE] Make sure callback finishes execution before running it the se…
pereanub Jun 11, 2026
5e07266
Update build requirement from <1.5 to <1.6 in /src/bindings/python (#…
dependabot[bot] Jun 11, 2026
4b0cb88
Update wheel requirement from <=0.45.1 to <=0.46.3 (#33785)
dependabot[bot] Jun 11, 2026
f884a3f
Bump kornia from 0.8.2 to 0.8.3 in /tests (#36008)
dependabot[bot] Jun 11, 2026
82c65a7
Bump urllib3 from 2.6.3 to 2.7.0 (#35730)
dependabot[bot] Jun 11, 2026
dae2a90
[SIT] Support YOLO outputs for mAP metric (#36261)
svigh Jun 11, 2026
7b32f06
[RT INFO] Switch from DisableFP16Compression to DisablePrecisionConve…
CuriousPanCake Jun 11, 2026
a0390c4
[CI] [GHA] Specify which file types are allowed as artefacts in CI Do…
akashchi Jun 12, 2026
81d8f2e
[ONNX FE] Add support for com.microsoft.MultiHeadAttention. (#36138)
nproshun Jun 12, 2026
c337f51
[GPU][test]: Refactoring of PA token type functional tests (#36277)
pkowalc1 Jun 12, 2026
de58f9a
Bump mistune from 3.2.0 to 3.2.1 in /docs (#35693)
dependabot[bot] Jun 12, 2026
cde2da7
Update pyright requirement from <=1.1.409 to <=1.1.410 in /src/bindin…
dependabot[bot] Jun 12, 2026
9b27af1
Bump beautifulsoup4 from 4.14.3 to 4.15.0 (#36292)
dependabot[bot] Jun 12, 2026
5519acc
[NPU] Fix spurious import rejection for page-adjacent host allocation…
ericcraw Jun 12, 2026
c91d326
MSVC 19.51 compiler bug WA (#36357)
Maxim-Doronin Jun 12, 2026
7cf8606
[GPU][Level Zero] Enable convolution tests for ZE on BMG (#36319)
Roszczyk Jun 12, 2026
510e5cc
[NPUW] Fix Qwen3-Embedding models for optimized attention modes (#36229)
AsyaPronina Jun 12, 2026
1722fcf
Expand MOE Support for Intel non-systolic GPUs
peterchen-intel Jun 13, 2026
470b534
Merge branch 'master' into oom/fixing
peterchen-intel Jun 13, 2026
4b0904d
[GPU] fix dGPU func testcases in smoke_MatMulCompressedWeights_extra_…
yuanxion Jun 14, 2026
58b4402
Merge branch 'master' into oom/fixing
peterchen-intel Jun 15, 2026
23b733b
Revert "[GPU] Fix implicit concat fusing blocked for feature-axis and…
davidsnam-intel Jun 15, 2026
816d605
[GPU] Move MoE Router related code to separate files (#36359)
v-Golubev Jun 15, 2026
ca82d24
Update comment
peterchen-intel Jun 15, 2026
3e141fc
Merge branch 'master' into oom/fixing
peterchen-intel Jun 15, 2026
b9a1f20
[CPU] TurboQuant codec for SDPA (#35853)
EgorDuplensky Jun 15, 2026
c88e94a
[NPUW][MoE Scaling]Qwen3-30B-A3B support on NPU. (#36269)
intelgaoxiong Jun 15, 2026
326f5ad
[CI] Shrink Android CPU test artifacts to fix CXX_Func_Tests timeout …
allnes Jun 16, 2026
50b57b5
[GPU] Fix paged_attention mixed-stage internal buffer layout mismatch…
sungeunk Jun 16, 2026
01f9f53
Bump js-yaml from 4.1.1 to 4.2.0 in /src/bindings/js/node (#36419)
dependabot[bot] Jun 16, 2026
4bca788
[CPU][Tests] Use member eltwise type in primitive selection (#36387)
aobolensk Jun 16, 2026
de92454
Added documentation for WinGet (#36324)
artanokhov Jun 16, 2026
a2b32c9
[GPU] Revert convolution fsv16 kernel changes because of OCL perf dro…
Roszczyk Jun 16, 2026
70cc71b
[GPU] Fix in-place crop padding leak with dynamic predecessor (#36420)
andrew-k-park Jun 16, 2026
8bd69db
[NPUW][MoE Performance]Overlap CPU and NPU workloads. (#36398)
intelgaoxiong Jun 16, 2026
81ec072
[Snippets][CPU] Unify GELU v0/v7 emitters across architectures (#36388)
aobolensk Jun 16, 2026
fcdfd38
[Transformations] Transformations utils: added extract_subgraph helpe…
v-Golubev Jun 16, 2026
1cb9fd3
[GPU][test]: PA token type unit tests refactoring. (#36327)
pkowalc1 Jun 16, 2026
d9ffee7
Bump pytest from 9.0.3 to 9.1.0 (#36402)
dependabot[bot] Jun 16, 2026
058b7cb
[GPU] Fix compilation failure for uncompressed (FP16) MoE models (#36…
andrew-k-park Jun 16, 2026
3d24f76
[GPU] support Gather-based RoPE pattern for position ids in LTX-Video…
ahnyoung-paul Jun 16, 2026
af9c5e6
[CPU][Tests] Skip InputNoReorderEltwiseBF16 on platforms without BF16…
aobolensk Jun 16, 2026
eed9db2
[Snippets][CPU][ARM] Enable weights repacking (#35531)
aobolensk Jun 16, 2026
a20c360
[CORE] parallel population for mmaped buffers (#35870)
almilosz Jun 16, 2026
a079fdd
[GPU] Add USM support to SYCL runtime (#36287)
Akihiro-Tabuchi Jun 16, 2026
e095c41
[CPU] PagedAttention: fixed cache collision (#36363)
v-Golubev Jun 16, 2026
65855d3
[core] Rename memory functions (#36296)
t-jankowski Jun 16, 2026
0508b91
Merge branch 'master' into oom/fixing
peterchen-intel Jun 17, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
22 changes: 0 additions & 22 deletions .claude/skills/debug/SKILL.md

This file was deleted.

Original file line number Diff line number Diff line change
@@ -1,6 +1,7 @@
---
name: debug-matcher-pass
description: Debug why an OpenVINO MatcherPass transformation is not firing. Use this skill immediately when a user says a transformation is "not applied", a "pass has no effect", a "matcher never triggers", a pattern "doesn't match", a "callback never fires", "WrapType predicate is too strict", a subgraph "not fused" despite the pass being registered, or they see "END: PATTERN DIDN'T MATCH" in matcher logs. Also trigger when a MatcherPass works on one model but silently skips another, when the user wants to add a reproducer test for a transformation that should fire but doesn't, or when they suspect an opset version mismatch preventing a match. Do NOT trigger for: writing a new MatcherPass from scratch, debugging a pass that fires but produces wrong numerical results, crashes in pass registration, or general questions about what MatcherPass is.
name: ov-debug-matcher-pass
description: >
Debug why an OpenVINO MatcherPass transformation is not firing. Use this skill immediately when a user says a transformation is "not applied", a "pass has no effect", a "matcher never triggers", a pattern "doesn't match", a "callback never fires", "WrapType predicate is too strict", a subgraph "not fused" despite the pass being registered, or they see "END: PATTERN DIDN'T MATCH" in matcher logs. Also trigger when a MatcherPass works on one model but silently skips another, when the user wants to add a reproducer test for a transformation that should fire but doesn't, or when they suspect an opset version mismatch preventing a match. Do NOT trigger for: writing a new MatcherPass from scratch, debugging a pass that fires but produces wrong numerical results, crashes in pass registration, or general questions about what MatcherPass is.
---

# Debug MatcherPass Skill
Expand All @@ -15,6 +16,8 @@ Produce two deliverables before finishing:

The only filesystem change should be the test file edit. Do not create additional output files.

**Early exit — all passes fired:** If Step 2a shows `CALLBACK SUCCEDED > 0` for every pass under investigation, skip Steps 3–6. Post a lightweight confirmation summary (use the "all fired" variant in the Diagnosis Report Template) and do not write a reproducer test. There is no bug to reproduce.

---

## Step 0: Gather Prerequisites
Expand All @@ -36,20 +39,20 @@ Ask the user only if the transformation name or run command is missing. The buil

## Step 1: Verify Debug Build

The matcher logging macros are compiled in only when `ENABLE_OPENVINO_DEBUG=ON`. Check whether the current build already has it:
The matcher logging macros are compiled in only when `ENABLE_DEBUG_CAPS=ON`. Check whether the current build already has it:

```bash
grep -i "ENABLE_OPENVINO_DEBUG" build/*/CMakeCache.txt
grep -i "ENABLE_DEBUG_CAPS" build/*/CMakeCache.txt
```

If the flag is absent or set to `OFF`, reconfigure CMake using the existing build directory and build type from Step 0:

```bash
cmake -B <build_dir> -DENABLE_OPENVINO_DEBUG=ON
cmake -B <build_dir> -DENABLE_DEBUG_CAPS=ON
cmake --build <build_dir> --parallel
```

> **Note:** Logging also works in `Release` builds as long as `ENABLE_OPENVINO_DEBUG=ON` is set at configure time.
> **Note:** Logging also works in `Release` builds as long as `ENABLE_DEBUG_CAPS=ON` is set at configure time.

---

Expand All @@ -68,6 +71,42 @@ Additional env vars:
- **`OV_MATCHERS_TO_LOG`** — comma-separated list of matcher names to filter (omit to log all matchers, which produces very large output).
- **`OV_VERBOSE_LOGGING=true`** — prints additional node details (element type, shape, attributes); use when the basic log does not identify the failure clearly.

### Wait for the command to complete

Matcher logs are interleaved with normal output. If the command is long-running (e.g., model compilation on GPU), **wait for it to finish before analyzing**. Parsing a partial log will produce incorrect statistics (e.g., undercounting successful callbacks). Verify completion by checking for the final your_run_command output line or the process exit code.

---

## Step 2a: Collect Per-Pass Statistics

Before diving into detailed log analysis, run a quick tally to build an overview of which passes fired and how many times. This is especially important for **pipeline cascade** investigations with multiple passes.

Use this Python snippet (adjust the `pass_names` list):

```bash
python3 -c "
import re, sys
pass_names = ['PassA', 'PassB'] # <-- replace with actual pass names
with open('matcher.log') as f:
content = f.read()
for p in pass_names:
s = len(re.findall(rf'\[{p}\] END: PATTERN MATCHED, CALLBACK SUCCEDED', content))
m = len(re.findall(rf'\[{p}\] END: PATTERN MATCHED', content))
d = len(re.findall(rf'\[{p}\] END: PATTERN DIDN.T MATCH', content))
print(f'{p}: CALLBACK SUCCEDED={s} MATCHED={m} DIDN\'T MATCH={d}')
"
```

Interpret the output:
- **CALLBACK SUCCEDED > 0** → pass fired and transformed nodes.
- **MATCHED > 0 but SUCCEDED = 0** → pattern matched but the callback returned `false` (check callback logic).
- **DIDN'T MATCH > 0 and MATCHED = 0** → pattern was attempted but never matched any node (proceed to Step 3 for root cause).
- **All counts = 0** → pass name not found in the log at all (not registered, or the pass is a `GraphRewrite` wrapper — log the inner `MatcherPass` names instead).

> **Note on `GraphRewrite` wrappers:** If a pass is a `GraphRewrite` that calls `add_matcher<InnerPass>()`, the matcher log will use the inner pass names, not the wrapper name. Check the header for inner pass names and use those in `OV_MATCHERS_TO_LOG`.

For pipeline cascades, use the tally to quickly classify each pass as ✅ fired / ❌ root cause / ❌ downstream casualty before spending time on detailed log analysis.

---

## Step 3: Analyze the Log — Identify Root Cause
Expand Down Expand Up @@ -128,12 +167,13 @@ Work inward through the nested blocks to find the deepest `}` labeled with a fai
7. **Attribute value mismatch** — Pattern constrains an attribute (e.g., `group == 1`, `axis == 0`) that doesn't match the actual node.
8. **Transformation was already applied** — Graph was modified by a symmetric or overlapping pass earlier; the target op no longer exists.
9. **Wrong opset version** — Pattern uses `opset::OpX` but the frontend or a prior pass has already replaced it with a different version or a decomposed form.
10. **Argument count mismatch** — The graph node has a different number of inputs than the pattern expects (e.g., an op with shared-expert inputs has 23 inputs but the pattern only covers the 11-input or 13-input variant). Common when ops have multiple configurations (with/without bias, with/without shared experts).

### 3e. No output at all from OV_MATCHER_LOGGING

If the log file is empty or contains no matcher output even though the flag is set:

- Confirm `ENABLE_OPENVINO_DEBUG=ON` in `CMakeCache.txt` for the binary you are actually running (not a different build directory).
- Confirm `ENABLE_DEBUG_CAPS=ON` in `CMakeCache.txt` for the binary you are actually running (not a different build directory).
- Confirm the env var is exported in the same shell context as the run command: `export OV_MATCHER_LOGGING=true` or prefix it inline.
- If calling through Python or a launcher script, the env var must survive into the child process — use `os.environ` or prefix the full command.
- Verify you are running the freshly rebuilt binary, not a cached one from a different `bin/` location.
Expand Down Expand Up @@ -280,11 +320,11 @@ See [references/example-diagnosis-report.md](references/example-diagnosis-report

<!-- If multiple passes were investigated, add this table: -->
## Summary of passes
| Pass | Result |
|---|---|
| `PassA` | ✅ Fired (`CALLBACK SUCCEEDED`) |
| `PassB` | ❌ Did not fire — root cause |
| `PassC` | ❌ Did not fire — downstream: PassB never produced its expected input |
| Pass | Result | Callbacks | Matches |
|---|---|---|---|
| `PassA` | ✅ Fired | 40 | 40 |
| `PassB` | ❌ Did not fire — root cause | 0 | 0 |
| `PassC` | ❌ Did not fire — downstream casualty | 0 | 0 |

**Root cause:** <one-sentence summary>
**Log evidence:** `<exact log phrase that identified the failure>`
Expand All @@ -297,6 +337,25 @@ Test name: `<TEST_F name>`
Status before fix: PASS (transformation did not fire — model unchanged, matches auto-cloned ref; confirms bug reproduced)
```

### All-fired variant (early exit)

When all passes fired successfully (Step 2a), use this shorter template instead:

```
## MatcherPass Diagnosis: <TransformationName(s)>

## Summary of passes
| Pass | Result | Callbacks | Matches |
|---|---|---|---|
| `PassA` | ✅ Fired | <N> | <N> |
| `PassB` | ✅ Fired | <N> | <N> |

**All transformations fired successfully.** No issues found.

## Reproducer Test
Not needed — no bug to reproduce.
```

---

## References
Expand Down
23 changes: 23 additions & 0 deletions .claude/skills/ov-debug/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,23 @@
---
name: ov-debug
description: Troubleshooting all sorts of failures, crashes, exceptions and errors using debug capabilities. Analyze accuracy, performance, model compilation, or memory issues. Dump tensors and intermediate blobs. Serialize and visualize IRs, execution graphs. Enable verbose, logging. Profile execution. Compare layer outputs. Inspect, trace or dump transformations. Identify executed operations, nodes, primitives, kernels.
---

# Debug Skill

## Prerequisites
Build flags that enable debug capabilities (check CMakeCache.txt in the build dir):
- `-DENABLE_DEBUG_CAPS=ON` — CPU/GPU plugin debug env vars, transformation matcher logging

## Components

| Component | Reference file to read | Routing hints |
|---------------------------|---------------------------------------|-------------------------------------------------------------------------------|
| openvino_intel_cpu_plugin | @components/debug-intel-cpu-plugin.md | CPU: inference issues, wrong results, slow inference, tensor dumps, execution graphs |
| openvino_intel_gpu_plugin | @components/debug-intel-gpu-plugin.md | GPU: inference issues, wrong results, slow inference, tensor dumps, execution graphs |
| transformations | @components/debug-transformations.md | transformation not applied, pass not firing, slow compilation, graph inspection |

## Steps
1. Match the user's symptom to the routing hints above to identify the component(s)
2. Load the component's reference file
3. Follow the instructions and recommendations for debugging
14 changes: 14 additions & 0 deletions .claude/skills/ov-debug/components/debug-transformations.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,14 @@
The debug and troubleshooting of OpenVINO transformations can be performed using various debug capabilities activated via environment variables.

# Reference
Read `src/common/transformations/docs/debug_capabilities/README.md` — use the "When to use" guidance to match the observed problem to the right capability.

# Handoff
If the symptom is specifically a MatcherPass not firing (transformation not applied, pattern not matching, callback never called), hand off to the `/ov-debug-matcher-pass` skill — it provides a deeper automated diagnosis workflow with matcher log analysis and reproducer test generation.

# Steps
1. Read the debug capabilities README
2. Match the observed symptom to the relevant "When to use" entries
3. If the issue is a MatcherPass not firing — invoke `/ov-debug-matcher-pass` instead of continuing
4. Read the linked detail doc for the chosen capability
5. Use suitable environment variables
28 changes: 28 additions & 0 deletions .claude/skills/ov-ensure-coding-style/SKILL.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
---
name: ov-ensure-coding-style
description: Detect and fix clang-format, clang-tidy, and copyright header violations in an OpenVINO C++ codebase. Use when the user complains about code style or formatting, asks to clean up changes, fix linting, add a copyright header, or when a style check or linting CI job is failing. Do not use for build errors, compilation failures, linker errors, test failures, runtime crashes, accuracy issues, or CMake config problems.
---

# Apply Code Standards

Iteratively detect and fix all clang-format, clang-tidy, and copyright violations introduced by the current branch's changes.

## Step 1: Identify Affected Files and Build Targets

**Before running any commands**, check whether the upstream reference is already confirmed in the conversation. If it is not, stop and ask:

> "I'll diff against `upstream/master`. Let me know if you use a different upstream."

Do not proceed until the user replies. Use the confirmed reference <ref_branch> in all subsequent steps. Once the upstream branch is known, you must fetch <ref_branch> and find a merge-base commit <ref_commit>, against which the diff should be collected. Then run:

```bash
git diff --name-only <ref_commit> | grep -v '^thirdparty' | tee /tmp/changed_files.txt
```

Map the changed files to CMake targets. If the mapping is not obvious, ask the user:

> "Which CMake targets cover the files you changed? (e.g., `openvino_intel_cpu_plugin`)"

## Step 2: Check, Fix, Repeat

**Follow the fix order and tool instructions from [coding_style.md](../../../docs/dev/coding_style.md) exactly — do not rearrange or skip steps.** Read that file before proceeding. Repeat the full cycle until all checks pass.
Loading
Loading