]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/shortlog
pkg/ggml/sources/whisper.cpp
2026-04-30 Georgi Gerganovggml : fix RWKV ops thread assignment (llama/21226)
2026-04-30 Taimur Ahmadggml-cpu: fix fallback for RVV kernels without zvfh...
2026-04-30 Anav PrasadCUDA: Add Flash Attention Support for Head Dimension...
2026-04-30 Reese Levineggml webgpu: quantized buffers to u32 + wider browser...
2026-04-30 Abhijit Rameshggml-webgpu: port all AOT operators to JIT (llama/20728)
2026-04-30 hipuddingCANN: fix multi-thread set_tensor race conditions ...
2026-04-30 Neo Zhangsycl : enhance fattn perf (llama/21185)
2026-04-30 shaofeiqiopencl: add q4_K gemm and gemv kernels for Adreno ...
2026-04-30 Oliver SimonsCUDA : Fix CUB's argsort when nrows % block_size =...
2026-04-30 Radoslav Gerganovrpc : fix misleading error log (llama/21184)
2026-04-30 Gaurav GargOptimize MOE GEMV kernel for BS > 1. (llama/20905)
2026-04-30 Max Krasnyanskyhexagon: dma optimizations (mostly fixing regressions...
2026-04-30 Georgi Gerganovggml : bump version to 0.9.9 (ggml/1449)
2026-04-20 jinweihanbench : sync submit-results URL to ggml-org (#3769)
2026-04-17 Daniel Worthington... whisper : add stateless VAD detect + explicit state...
2026-03-29 Georgi Gerganovsync : ggml
2026-03-29 Ruben Ortlamvulkan: add noncontiguous GLU support (llama/21081)
2026-03-29 Yiwei Shaohexagon: support for IQ4_NL and MXFP4 (llama/21018)
2026-03-29 Radoslav Gerganovrpc : proper handling of data pointers to CPU buffers...
2026-03-29 renmetal : Fix dimension constraint violation in matmul2d...
2026-03-29 uvoship: use fnuz fp8 for conversion on CDNA3 (llama/21040)
2026-03-29 lhezopencl: allow large buffer for adreno (llama/20997)
2026-03-29 ihb2032fix(ggml): correct RISC-V ISA string canonical ordering...
2026-03-29 Michael Wandggml-cuda: Add NVFP4 dp4a kernel (llama/20644)
2026-03-29 Yihao WangCUDA & CPU: support F32 kernel type for `CONV_TRANSPOSE...
2026-03-29 Saba Fallahmtmd: Add DeepSeekOCR Support (llama/17400)
2026-03-29 Johannes Gäßlerllama: fix llama-model-saver (llama/20503)
2026-03-29 Neo Zhangsycl : fix wrong variable check by assert (llama/20903)
2026-03-29 nurimetal : add FLOOR, CEIL, ROUND, TRUNC unary ops (llama...
2026-03-29 Georgi Gerganovmetal : add FA instantiations for HSK=512, HSV=512...
2026-03-29 Max Krasnyanskyhexagon: general DMA and Binary Op fixes for large...
2026-03-29 lhezopencl: add q6_K gemm and gemv kernels for Adreno ...
2026-03-29 las7rpc : RCE patch (llama/20908)
2026-03-29 Rashid Ul Islammetal: add CONV_3D (llama/19927)
2026-03-29 Chenguang LiCANN: add RoPE cache preload before ACL graph capture...
2026-03-29 Dan Hoffmanfix(openvino): explicit memset in buffer_context alloca...
2026-03-29 shaofeiqiopencl: add flattened Q4_K mv and general Q4_K mm ...
2026-03-29 Johannes GäßlerCUDA: fix BF16 FA compilation (llama/20865)
2026-03-29 Neo Zhangsupport bf16 and quantized type (llama/20803)
2026-03-29 Patrick Buckleyggml-cuda: native bf16 flash attention for vec kernel...
2026-03-29 Gaurav GargIncrease number of output elements per-thread block...
2026-03-29 y198fix(rpc): prevent division by zero in deserialize_tenso...
2026-03-29 Matt CoralloAdd shader count for Intel Arc Pro B60 (llama/20818)
2026-03-29 shalinib-ibmggml-cpu: add always_inline to tinyBLAS_PPC accumulator...
2026-03-29 Jeff Bolzvulkan: change gated_delta_net to shard a column across...
2026-03-29 hipuddingCANN: add BF16 support for core operators (llama/20152)
2026-03-29 Sundaram krishnanggml: guard KleidiAI DOWNLOAD_EXTRACT_TIMESTAMP for...
2026-03-29 Rail Chabdarovhip: Avoid compiler bug in RDNA code generation during...
2026-03-29 Yiwei Shaohexagon: add Matrix Extensions (HMX) for Hexagon NPU...
2026-03-29 uvosci : add hip quality check (llama/20430)
2026-03-29 Reese Levineggml webgpu: ops support for qwen3.5 (SET, TRI_SOLVE...
2026-03-29 Evevulkan: dequantize iq4_xs 4 at a time (llama/20657)
2026-03-29 Charles Xucmake : fix build warning when kleidiai is enabled...
2026-03-29 Chenguang LiCANN: handle in-place ROPE on non-contiguous f32 tensor...
2026-03-29 Masashi Yoshimuraggml-webgpu: Update the `RMS_NORM` preprocessor and...
2026-03-29 Masashi Yoshimuraggml-webgpu: Add supports for `DIAG` and `TRI` (llama...
2026-03-29 Chenguang LiCANN: support flash attention for head dim not multiple...
2026-03-29 Reese LevineMove to no timeout for WaitAny in graph submission...
2026-03-29 Shaw Nguyenggml-cpu/x86: fix unused changemask warning in repack...
2026-03-29 uvosHIP : ignore return of hipMemAdvise [no ci] (llama...
2026-03-29 Krishna Sridharhexagon: add neg, exp, sigmoid, softplus ops, cont...
2026-03-29 Ruben Ortlamvulkan: disable mmvq on Intel Windows driver (llama...
2026-03-29 Kevin Hannonggml-blas: set mkl threads from thread context (llama...
2026-03-29 Taimur Ahmadggml-cpu: fix RVV checks in quants and repacking (llama...
2026-03-29 Ruben Ortlamvulkan: async and event fixes (llama/20518)
2026-03-29 Justin Bradfordkleidiai : fix MUL_MAT support for batched (3D) inputs...
2026-03-29 Ruben Ortlamvulkan: allow graphics queue only through env var ...
2026-03-29 Neo Zhangehance UPSCALE to support all UT cases (llama/20637)
2026-03-29 Martin Klacerkleidiai: add data type check to get_tensor_traits...
2026-03-29 Ruben Ortlamvulkan: fix flash attention dot product precision ...
2026-03-29 Aman GuptaCUDA: GDN hide memory latency (llama/20537)
2026-03-29 Sigbjørn Skjæretsycl : fix for untransposed GDA recurrent state (llama...
2026-03-21 KITAITI Makotoruby : fix dangling pointers, memory leak, and SEGV...
2026-03-19 Georgi Gerganovrelease : v1.8.4 upstream/1.8.4
2026-03-18 Georgi Gerganovci : update workflows
2026-03-18 Georgi Gerganovbenches : update
2026-03-18 Georgi Gerganovsync : ggml
2026-03-18 Georgi Gerganovggml : bump version to 0.9.8 (ggml/1442)
2026-03-18 Georgi Gerganovggml : restore ggml_type_sizef() to aboid major version...
2026-03-17 lohopupafix: VAD time mapping timestamp drift caused by overlap...
2026-03-16 Alango : handle EOF correctly in model download (#3671)
2026-03-16 Aiudadadadfpy : replace deprecated openvino-dev with openvino...
2026-03-16 Gaël Jamesexamples : Allow max_len to be used for any output...
2026-03-16 Igor Loskutovserver: return proper HTTP status codes for error respo...
2026-03-16 Georgi Gerganovggml : try fix arm build (#0)
2026-03-16 Georgi Gerganovtalk-llama : sync llama.cpp
2026-03-16 Georgi Gerganovsync : ggml
2026-03-16 David366AIggml : extend im2col f16 (ggml/1434)
2026-03-16 Georgi Gerganovcommon : add nvfp4 (ggml/0)
2026-03-16 Johannes GäßlerCUDA: limit number of FA stream-k CUDA blocks (llama...
2026-03-16 Pascalggml: avoid creating CUDA context during device init...
2026-03-16 MoonShadowggml/hip: fix APU compatibility - soft error handling...
2026-03-16 Bartowskiggml : guard against sumq2 being 0 in IQ4_NL (llama...
2026-03-16 PikaPikachucuda : add RDNA4-specific MMVQ parameter table for...
2026-03-16 Ruben Ortlamvulkan: use graphics queue on AMD (llama/20551)
2026-03-16 Georgi Gerganovmetal : add FA specialization for HSK = 320, HSV =...
2026-03-16 Max Krasnyanskyhexagon: Q4_0 and MXFP4 repack fixes (llama/20527)
2026-03-16 Neo Zhangadd op gated_delta_net (llama/20455)
2026-03-16 Adrien Gallouëtggml : add native AVX512-FP16 support for F16 operation...
2026-03-16 WallentriUse fp32 in cuBLAS V100 to avoid overflows, env variabl...
next