]> git.djapps.eu Git - pkg/ggml/sources/ggml/shortlog
pkg/ggml/sources/ggml
2025-01-03 Srihari-mcwggml : fixes for AVXVNNI instruction set with MSVC...
2025-01-03 Jeff Bolzvulkan: optimize mul_mat for small values of N (llama...
2025-01-03 Jeff Bolzvulkan: im2col and matmul optimizations for stable...
2025-01-03 Jeff Bolzvulkan: Use push constant offset to handle misaligned...
2025-01-03 Evevulkan: multi-row k quants (llama/10846)
2025-01-03 Peterexamples, ggml : fix GCC compiler warnings (llama/10983)
2025-01-03 Djip007ggml : more perfo with llamafile tinyblas on x86_64...
2025-01-03 Diego Devesaggml : use wstring for backend search paths (llama...
2025-01-03 Diego Devesaggml : fix arm enabled features check (llama/10961)
2025-01-03 Diego Devesaggml : fix const usage in SSE path (llama/10962)
2025-01-03 yuri@FreeBSDggml : fix run-time on FreeBSD in get_executable_path...
2025-01-03 Jeff Bolzvulkan: build fixes for 32b (llama/10927)
2025-01-03 Jeff Bolzvulkan: optimize coopmat2 dequant functions (llama...
2025-01-03 Adrien Gallouëtggml-cpu: replace NEON asm with intrinsics in ggml_gemv...
2025-01-03 Akarshan BiswasSYCL: Migrate away from deprecated ggml_tensor->backend...
2025-01-03 Diego Devesaggml : add test for SVE and disable when it fails ...
2025-01-03 Adrien Gallouëtggml: fix arm build with gcc (llama/10895)
2025-01-03 Diego Devesaggml : fix arm build (llama/10890)
2025-01-03 Georgi Gerganovtts : add OuteTTS support (llama/10784)
2025-01-03 Johannes Gäßlertests: add tests for GGUF (llama/10830)
2024-12-23 Georgi Gerganovggml : do not install metal source when embed library...
2024-12-19 Daniel Beveniusggml : improve inputs log sched_print_assignments ...
2024-12-18 Georgi Gerganovsync : whisper.cpp
2024-12-18 Georgi Gerganovfiles : remove old sources (#1052)
2024-12-17 Georgi Gerganovsync : llama.cpp
2024-12-17 Georgi Gerganovggml : update ggml_backend_cpu_device_supports_op ...
2024-12-17 Evevulkan: bugfixes for small subgroup size systems +...
2024-12-17 Zhiyuan Lirwkv6: add wkv6 support for Vulkan backend (llama/10829)
2024-12-17 HimariOllama : add Qwen2VL support + multimodal RoPE (llama...
2024-12-17 lhezIntroducing experimental OpenCL backend with support...
2024-12-17 谢乃闻Fix crash caused by ggml_backend_load_all when launchin...
2024-12-17 Evevulkan: small mul_mat_vec optimizations (llama/10665)
2024-12-17 Akarshan BiswasSYCL: Reduce most of the compiler warnings (llama/10748)
2024-12-17 Karol Kontnyggml : Fix compilation issues on ARM platform when...
2024-12-17 a3shCUDA: faster non-contiguous concat (llama/10760)
2024-12-17 Diego Devesaremove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS (llama/10797)
2024-12-17 0cc4mVulkan: Use improved q4_k and q5_k dequant code in...
2024-12-17 0cc4mVulkan: Add VK_EXT_subgroup_size_control support to...
2024-12-17 Gilad Sggml: load all backends from a user-provided search...
2024-12-17 Jeff Bolzvulkan: request round-to-even for fp16 in im2col/rope_h...
2024-12-17 Evevulkan: dynamic subgroup size for the remaining k quant...
2024-12-17 Andreas KieslingerCUDA: rename macros to avoid conflicts with WinAPI...
2024-12-17 Jeff Bolzvulkan: disable spirv-opt for coopmat shaders (llama...
2024-12-17 Georgi Gerganovsync : whisper.cpp
2024-12-17 Georgi Gerganovcmake : fix "amd64" processor string (whisper/2638)
2024-12-17 gn64vulkan : fix soft_max.comp division by zero (whisper...
2024-12-14 Daniel Beveniusggml : remove return from ggml_gallocr_allocate_node...
2024-12-13 Daniel Beveniusggml : add check for grad_accs (#1046)
2024-12-10 Georgi Gerganovcommon : remove old types
2024-12-10 Georgi Gerganovsync : llama.cpp
2024-12-10 Johannes GäßlerCUDA: fix shared memory access condition for mmv (llama...
2024-12-10 Jeff Bolzvulkan: fix compile warnings (llama/10731)
2024-12-10 stduhpfVulkan: fix NaN in tanh.comp with AMD proprietary drive...
2024-12-10 Jeff Bolzvulkan: compile a test shader in cmake to check for...
2024-12-10 Georgi Gerganovggml : disable iq4_nl interleave size 8 (llama/10709)
2024-12-10 Djip007ggml : refactor online repacking (llama/10446)
2024-12-10 0cc4mVulkan: VK_KHR_cooperative_matrix support to speed...
2024-12-10 Robert Ormandimetal : Extend how Llama.cpp locates metal resources...
2024-12-10 Jeff Bolzvulkan: Add VK_NV_cooperative_matrix2 support for mul_m...
2024-12-08 Georgi Gerganovsync : whisper.cpp
2024-12-08 Georgi Gerganovscripts : update whisper sync
2024-12-05 Georgi Gerganovsync : llama.cpp
2024-12-05 Diego Devesaggml : add predefined list of CPU backend variants...
2024-12-05 Diego Devesaggml-cpu : fix HWCAP2_I8MM value (llama/10646)
2024-12-05 Jeff Bolzvulkan: Implement "fast divide" (mul+shift) for unary...
2024-12-05 Nicolò ScipioneSYCL : Move to compile time oneMKL interface backend...
2024-12-05 Frankie RobertsonAvoid using __fp16 on ARM with old nvcc (llama/10616)
2024-12-05 Jeff Bolzvulkan: optimize and reenable split_k (llama/10637)
2024-12-04 PABggml: add `GGML_SET` Metal kernel + i32 CPU kernel...
2024-12-03 PABggml : add `GGML_PAD_REFLECT_1D` operation (#1034)
2024-12-03 Georgi Gerganovci : update requirements.txt
2024-12-03 Georgi Gerganovci : remove opencl workflow
2024-12-03 Georgi Gerganovci : fix pip env
2024-12-03 Georgi Gerganovfiles : remove make artifacts
2024-12-03 Georgi Gerganovcommon : fix compile warning
2024-12-03 Georgi Gerganovauthors : update
2024-12-03 Georgi Gerganovsync : llama.cpp
2024-12-03 Diego Devesaggml : move AMX to the CPU backend (llama/10570)
2024-12-03 Georgi Gerganovscripts : remove amx from sync
2024-12-03 Georgi Gerganovmetal : small-batch mat-mul kernels (llama/10581)
2024-12-03 Akarshan BiswasSYCL: Fix and switch to GGML_LOG system instead of...
2024-12-03 Adrien Gallouëtggml-cpu: replace AArch64 NEON assembly with intrinsics...
2024-12-03 Evevulkan: Dynamic subgroup size support for Q6_K mat_vec...
2024-12-03 Georgi Gerganovggml : fix I8MM Q4_1 scaling factor conversion (llama...
2024-12-03 Shupei Fanggml-cpu: fix typo in gemv/gemm iq4_nl_4_4 (llama/10580)
2024-12-03 Alberto Cabrera... sycl : offload of get_rows set to 0 (llama/10432)
2024-12-03 Alberto Cabrera... sycl : Reroute permuted mul_mats through oneMKL (llama...
2024-12-03 Chenguang LiCANN: RoPE operator optimization (llama/10563)
2024-12-03 Jeff Bolzvulkan: get the first command buffer submitted sooner...
2024-12-03 Georgi Gerganovggml : remove redundant copyright notice + update authors
2024-12-03 Georgi Gerganovggml : fix row condition for i8mm kernels (llama/10561)
2024-12-03 Georgi Gerganovcmake : fix ARM feature detection (llama/10543)
2024-12-03 Shupei Fanggml-cpu: support IQ4_NL_4_4 by runtime repack (llama...
2024-12-03 Sergio Lópezkompute : improve backend to pass test_backend_ops...
2024-12-03 leo-ponyCANN: Fix SOC_TYPE compile bug (llama/10519)
2024-12-03 Chenguang LiCANN: ROPE operator optimization (llama/10540)
2024-12-03 uvosAdd some minimal optimizations for CDNA (llama/10498)
2024-12-03 Georgi Gerganovmetal : fix group_norm support condition (llama/0)
2024-12-03 Jeff Bolzvulkan: define all quant data structures in types.comp...
2024-12-03 Jeff Bolzvulkan: Handle GPUs with less shared memory (llama...
next