Commits openvinotoolkit/openvino

[TRANSFORMATIONS] Fixed RoPEFusionQwen transformation (#31720)

mryzhov committed 275 days ago

Verified d8117c61

Bump product version 2025.4 (#31737)

akladiev committed 275 days ago

Verified 46daa2d4

Bump github/codeql-action from 3.29.4 to 3.29.5 (#31522)

dependabot[bot] committed 276 days ago

Verified 7cc4b1a5

[GPU] Fix coverity issue (#31723)

davidsnam-intel committed 276 days ago

Verified 689ab8b0

NPUW: Precompute and store sin/cos ops in ROPE in fp32 (#31644)

esmirno committed 276 days ago

Verified efcb061e

[Protopipe] Add output clamping option for OpenVINO models (#31713)

Andrei Tamas committed 276 days ago

Verified 51b2bac2

[NPU] Add support for QDQ_OPTIMIZATION_AGGRESSIVE compiler property (#31686)

pereanub committed 276 days ago

Verified 16019721

[Snippets][CPU] Enable Matmul with transposed B matrix tests (#31706)

aobolensk committed 276 days ago

Verified 73d64a87

rewrite StatefulSDPAFusion transformation using new symbolic API (#31709)

evkotov committed 277 days ago

Verified bfc0ee2b

[GPU] Fix nan values in SoftPlus (#31654)

davidsnam-intel committed 277 days ago

Verified 890c0583

[GHA] Don't store nightly build artifacts in cloud (#31712)

akladiev committed 277 days ago

Verified baf2ffa1

[NPUW] WA for second token generation when weight bank for generate is on CPU (#31703)

AsyaPronina committed 277 days ago

Verified 7ad84da0

[TRANSFORMATIONS][GPU] Fix bool type of attention mask decomposition issue in ScaledDotProductAttentionDecomposition (#31704)

andrew-k-park committed 277 days ago

Verified cca65cf3

Add support of u2 type for Subtract op in template plugin (#31645)

olpipi committed 277 days ago

Verified 84c08430

[CI] [GHA] Add `OperationStopped:` to known errors (#31711)

akashchi committed 277 days ago

Verified 4c938b09

[Snippets][Docs] Aligned Snippets documentation with the current state (#31680)

a-sidorova committed 277 days ago

Verified 28880f53

Fix "Overflowed integer argument (INTEGER_OVERFLOW)" Coverity remarks (#31705)

chenhu-wang committed 277 days ago

Verified cb26ee3e

[NPUW] Extend optimize_v_tensor patterns with optional FakeConvert (#31642)

smirnov-alexey committed 277 days ago

Verified d05f47f9

[IE-MDK][NPU] Enable canRun3RequestsConsistently tests (#31653)

sbutnari committed 277 days ago

Verified b03e7d57

[NPU] Refactor the igraph class (#31652)

pereanub committed 277 days ago

Verified 6876e8c0

[Snippets][CPU] Fix stride issues in jit_gemm_copy_b_emitter (#31601)

aobolensk committed 277 days ago

Verified dacd71e7

Transformations Infrastructure: MultiMatcher (#30587)

itikhono committed 278 days ago

Verified 652e211a

Add tbb new interface set_core_types() call which support P cores and E cores binding on PTL (#31631)

sunxiaoxia2022 committed 278 days ago

Verified 58bf2f93

NPUW: Automatically align MAX_PROMPT_LENGTH to CHUNK_SIZE (#31687)

dmatveev committed 278 days ago

Verified 02188dab

E2e accuracy improvement (#31678)

isanghao committed 278 days ago

Verified c16e0d63

[NPU] Check if set shape is out of bounds (#31685)

pereanub committed 278 days ago

Verified 33a3c446

[PT FE] Support aten.nonzero and short strides for convolution (#31684)

mvafin committed 278 days ago

Verified 16159d35

[Snippets][CPU] Optimize runtime offset handling in ARM64 memory emitters (#31665)

aobolensk committed 278 days ago

Verified 70ac4578

[Snippets][CPU] Fix handling negative numbers in Gemm executor, adjust its testing scenarios (#31636)

aobolensk committed 278 days ago

Verified 788f452e

[Snippets][CPU] Optimize runtime offset handling in ARM64 kernel emitters and utils (#31668)

aobolensk committed 278 days ago

Verified 21c2b7be