{"full_name":"KhronosGroup/MoltenVK","owner":"KhronosGroup","name":"MoltenVK","shard":"1","url":"https://github.com/KhronosGroup/MoltenVK","description":"MoltenVK is a Vulkan Portability implementation. It layers a subset of the high-performance, industry-standard Vulkan graphics and compute API over Apple's Metal graphics framework, enabling Vulkan applications to run on macOS, iOS and tvOS.","homepage":"","topics":[],"tag_pages":[],"language":"Objective-C++","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/1608701?v=4","created_at":"2017-11-07T18:41:56Z","pushed_at":"2026-10-13T17:33:45Z","metrics":{"stars":5933,"forks":609,"watchers":128,"issues":1256,"prs":1533,"releases":80,"issues_open":326,"issues_closed":930,"prs_open":58,"prs_closed":89,"prs_merged":1386,"discussions":88},"growth":{"day":{"stars":5,"forks":1,"base_day":"2026-11-15"}},"history":{"dates":["2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[5921,5922,5925,5926,5928,5928,5933],"forks":[605,606,606,606,607,608,609],"watchers":[null,128,128,128,128,128,128],"issues_open":[null,326,326,326,326,326,326],"prs_open":[null,57,57,57,58,58,58]},"boards":{},"same_owner":[],"issues":[{"number":203,"title":"Requirements for DXVK support..","comments":174,"created_at":"2018-07-27T06:58:46Z"},{"number":2054,"title":"Detroit: Become Human crashes under Wine","comments":153,"created_at":"2023-11-01T21:31:37Z"},{"number":427,"title":"Raytracing support","comments":53,"created_at":"2018-12-26T00:13:56Z"},{"number":1542,"title":"Flickering in highly lit areas on Mac in FFXIV","comments":41,"created_at":"2022-03-09T14:19:39Z"},{"number":1524,"title":"Add support for geometry shaders","comments":41,"created_at":"2022-02-11T11:03:16Z"},{"number":214,"title":"VKD3D Vulkan requirements..","comments":36,"created_at":"2018-08-04T18:03:44Z"},{"number":1650,"title":"FFXIV fails to run on Mac with > 100GB RAM","comments":26,"created_at":"2022-07-16T23:28:38Z"},{"number":1507,"title":"Dark Souls 3: Unstable frame rate on macOS","comments":24,"created_at":"2022-01-17T12:37:57Z"}],"discussions":[{"number":2751,"title":"can one Compile debug-mode libMoltenVK.dylib by OCLP intel-mac/Ventura; for sole purpose of Running on true-blue intel-mac Monterey","comments":1,"category":"Q&A","created_at":"2026-06-06T22:56:27Z"},{"number":2716,"title":"MoltenVK Limitations on iOS Simulator","comments":0,"category":"Q&A","created_at":"2026-04-16T11:18:54Z"},{"number":2709,"title":"Congrats for the latest Minecraft snapshot release","comments":0,"category":"General","created_at":"2026-04-07T16:18:12Z"},{"number":2688,"title":"Investigating performance of MSAA","comments":1,"category":"Q&A","created_at":"2026-02-05T07:32:45Z"},{"number":2633,"title":"How do I build MoltenVK with private api enabled?","comments":1,"category":"Q&A","created_at":"2025-10-08T01:53:41Z"},{"number":2555,"title":"Does MoltenVK allow vertexAttributeAccessBeyondStride if the stride is zero?","comments":1,"category":"Q&A","created_at":"2025-06-02T18:21:47Z"},{"number":2552,"title":"Specialization constants and local group sizes","comments":0,"category":"General","created_at":"2025-05-29T15:58:41Z"},{"number":2540,"title":"When do you expect the official VulkanSDK to take MVK 1.3+ ?","comments":1,"category":"Q&A","created_at":"2025-05-13T22:44:47Z"}],"latest_release":{"name":"Release v1.4.2","tag":"v1.4.2","published_at":"2026-07-24T14:00:46Z","body":"- Add support for the following extensions:\r\n  - `VK_EXT_sampler_filter_minmax` on Apple10 GPUs with macOS 26 or iOS 26\r\n- Add gl_DrawID / DrawIndex support for MacOS.\r\n- Add native sampler min/max filtering.\r\n- Fix buffer and heap out-of-sync in `initExternalMemory()`.\r\n- Fix incorrect varable usage in `MVKImagePlane::getMTLTexture()`.\r\n- Fix incorrect placement heap check in `MVKImagePlane::getMTLTexture()`.\r\n- Fix device loss when using imported `MTLTexture` objects with Metal argument buffers.\r\n- Raise minimum target to _macOS 12.0 / iOS 15.0 / tvOS 15.0_\r\n- Fix desync in primitive restart enable state tracking.\r\n- Fix inverted return logic in MVKBitArray::areAnyBitsSet().\r\n- Fix imported MTLTexture residency tracking.\r\n- Fix swapchain recreation giving 1x1 drawables.\r\n- Fix variable descriptor count query.\r\n- Fix channel corruption on Apple Silicon for color RTs used as transfer sources.\r\n- Fix behavior of VK_REMAINING_ARRAY_LAYERS with 2D array views of 3D images.\r\n- Fix invalidation of cached metal textures\r\n- Fix AMD subgroup size (64→32) and disable Intel UHD 630 simdReduction.\r\n- Fix incorrect placement heap check in `MVKImagePlane::getMTLTexture()`.\r\n- Fix regressions and crashes on Mac 1 (NVIDIA/Intel) GPUs.\r\n- Disable 64-bit float control properties.\r\n- Disable non-working quad control flow.\r\n- Emulate reversed-depth viewports on affected AMD Mac GPUs.\r\n- Do not rely on extension enable for vkGetPhysicalDeviceToolPropertiesEXT procaddr.\r\n- Do not convert UINT8 indices until draw time.\r\n- Do not enable blending when unsupported by format.\r\n- Null-check pColorAttachmentInputIndices before reading.\r\n- Make imported `MTLBuffer` resident in `MVKDeviceMemory`.\r\n- Include Metal constant buffer alignment in argument buffer gpuAlign calculation.\r\n- Use max of pointer and constant buf alignments for arg buffer uniform.align field.\r\n- Only apply Metal Constant Buffer alignment to Aux Buffers and Combined Textures.\r\n- Apply Metal constant buffer alignment padding for each descriptor set.\r\n- MVKRenderPass: encode barriers for subpass dependencies.\r\n- Occlusion queries: wrap detection uses stale baseline, crashes when visibility buffer starts near end of pass.\r\n- Prevent null dereference in MVKPixelFormats.\r\n- Improve handling of draw ID buffer binding.\r\n- Restore some checks for Mac1 GPUs.\r\n- Update copyright notices to year 2026.\r\n- Update to latest SPIRV-Cross:\r\n  - MSL: add `gl_DrawID` (DrawIndex) support for macOS.\r\n  - MSL: Add reversed depth viewport emulation.\r\n  - MSL: Fix `subgroupBallotExclusiveBitCount()` is not available for task shader and mesh shader.\r\n  - MSL: `thread_execution_width` is deprecated as of Metal 3.0 , use `threads_per_simdgroup` instead.\r\n  - MSL: Add `msl3.0` variant for subgroup builtin.\r\n  - MSL: Fix handling of variable pointer arguments which are not BDA.\r\n  - MSL: Fix variable pointer buffer arguments.\r\n  - MSL: Handle variable pointer argument to SSBO array.\r\n  - MSL: Fix handling of forced temporary row-major matrices.\r\n  - MSL: Additional MSL specific keywords(`signed`).\r\n  - MSL: Deal with std140 layouts when copying arrays between storage class.\r\n  - MSL: Rewrite std140 struct padding.\r\n  - MSL: BlockIO in vertex is captured as thread storage.\r\n  - MSL: Add initial cooperative matrix support.\r\n  - MSL: Fix ptr-cast prepass and coopmat typed load/store.\r\n  - MSL: Handle array type writes to builtin clip/cull better.\r\n  - MSL: Fix some extremely bizarre tess access chain bugs.\r\n  - MSL: Use pointer expression when dealing with atomics.\r\n  - Parser: support SPIR-V library modules without OpEntryPoint","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}