]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2026-05-29 Xuan-Son Nguyenserver: bump timeout to 3600s (#23842)
2026-05-29 fairydreamingmodel : support for DeepseekV32ForCausalLM with generic...
2026-05-29 Aman Guptallama: use f16 mask for FA to save VRAM (#23764)
2026-05-29 Georgi Gerganovsync : ggml
2026-05-29 Georgi Gerganovggml : bump version to 0.13.1 (ggml/1523)
2026-05-29 Omid Azizingram-mod : Add missing include (#23857)
2026-05-29 Aman Guptallama: add llm_graph_input_mtp (#23643)
2026-05-29 Adrien Gallouëtapp : move licences to llama-app (#23824)
2026-05-29 Andreas Kieslingercuda : disables launch_fattn PDL enrollment due to...
2026-05-29 Matt Corallometa : Add missing `buffer` set in allreduce fallback...
2026-05-28 Max Krasnyanskyhexagon: basic/generic op fusion support and RMS_NORM...
2026-05-28 Xuan-Son Nguyenmtmd-debug: add color and rainbow mode (#23829)
2026-05-28 Xuan-Son Nguyenmtmd: fix gemma 4 projector pre_norm (#23822)
2026-05-28 lhezopencl: move backend info printing into its own functio...
2026-05-28 Sigbjørn Skjæretci : run ui publish on ubuntu-slim (#23818)
2026-05-28 ValdikSSui: fix audio and video modality detection (#23756)
2026-05-28 Georgi Gerganovci : releases use Github-hosted builds for the UI ...
2026-05-28 Adrien Gallouëtapp : improve help output (#23805)
2026-05-28 Saba Fallahmtmd: n_head_kv defaults to n_head (#23782)
2026-05-28 Xuan-Son Nguyenmtmd: fix gemma 4 audio rms norm eps (#23815)
2026-05-28 Georgi Gerganovci : change Vulkan builds to Release to reduce ccache...
2026-05-28 Mikolaj Kucharskiarg: Add LLAMA_ARG_API_KEY_FILE environment variable...
2026-05-28 Johannes Gäßlertest-llama-archs: fix table format [no release] (#23810)
2026-05-28 fl0rianrggml: auto apply iGPU flag CUDA/HIP if integrated devic...
2026-05-28 redfoxmmvq Optim: add MMVQ_PARAMETERS_TURING(mmvq_parameter_...
2026-05-28 Jaden_MachCUDA: route batch>=4 quantized matmul to MMQ on AMD...
2026-05-28 Funtowicz Morganserver: minor tweaks to use more cpp features (#23785)
2026-05-28 Max Krasnyanskyhexagon: minor refresh for HMX FA and MM (#23796)
2026-05-28 Jeff Bolzvulkan: fast path for walsh-hadamard transform (#23687)
2026-05-28 Jesus Talaverachat : add Granite 4.1 chat template (#23518)
2026-05-28 Winston Mavulkan: fix wrong index variable in inner loop (#23665)
2026-05-28 Winston Mavulkan: Fix memory logger unsafe iterator access (...
2026-05-28 Markus Tavenrathserver, ui : Add support for HTTP ETags in llama-server...
2026-05-28 Sachin Sharmadocker : add ZenDNN Dockerfile (#23716)
2026-05-28 fairydreamingcuda : fix KQ mask offset integer overflow in fattn...
2026-05-28 Adrien Gallouëtperplexity : fix format specifier in LOG_ERR (#23788)
2026-05-28 ynankaniconvert : add FP8 to Q8 conversion (#23250)
2026-05-28 Martin Klacerggml: fixed Arm SVE usage bug in vec.h, vec.cpp (#22841)
2026-05-28 Georgi Gerganovci : refactor (#23789)
2026-05-28 ymckiHexagon: OP_GATED_DELTA_NET K>1 support (#23531)
2026-05-28 ymckiopencl: OP_GATED_DELTA_NET (#23312)
2026-05-27 Reese Levineggml-webgpu: remove legacy constants (#23672)
2026-05-27 Max Krasnyanskyhexagon: add support for Q4_1 in MUL_MAT and MUL_MAT_ID...
2026-05-27 Masashi Yoshimuraggml-webgpu: Fix how to dispatch WG to some ops (#23750)
2026-05-27 Matt Corallovulkan: Switch MUL_MAT_VEC to 4 K per iteration for...
2026-05-27 Jeff Bolzvulkan: use GL_NV_cooperative_matrix_decode_vector...
2026-05-27 l8bloomvulkan: add REPEAT op support for f16 to f16. (#23298)
2026-05-27 Georgi Gerganovci : move ARM jobs to self-hosted + disable kleidiai...
2026-05-27 Alessandro... vendor : update cpp-httplib to 0.46.0 (#23650)
2026-05-27 Sigbjørn Skjæretpyproject : add conversion folder and update dependenci...
2026-05-27 Oliver SimonsCUDA: restrict PDL to CTK >= 12.3 due to MSVC issues...
2026-05-27 Sigbjørn Skjæretci : bump cuda release to 13.3 (#23749)
2026-05-27 Georgi Gerganovcommon : fix env names to all have LLAMA_ARG_ prefix...
2026-05-27 Georgi Gerganovci : fix windows ccaches (#23777)
2026-05-27 Sigbjørn Skjæretci : remove wasm test (#23733)
2026-05-27 Winston Mavulkan: avoid preferring transfer queue on AMD UMA...
2026-05-27 Georgi Gerganovci : add ccache to server builds + fix undefined saniti...
2026-05-27 quyentonndbsdocs : fix duplicated "the" in granitevision and model...
2026-05-27 zhangtao2-1convert: add MiniCPM5 tokenizer support (#23384)
2026-05-27 Radoslav Gerganovserver : fix the log message when using SSL (#23393)
2026-05-26 Vladislavggml-zendnn : fixed naming of matmul function (#20964)
2026-05-26 Georgi Gerganovci : do not allocate ccache for 3rd-party hosted runner...
2026-05-26 Georgi Gerganovci : move [no release] check to dedicated check_release...
2026-05-26 Georgi Gerganovci : add `[no release]` keyword + fix sanitizer builds...
2026-05-26 Georgi Gerganovci : move macos jobs to the apple workflow + fix names...
2026-05-26 Jeff Bolzvulkan: optimize conv2d and implement coopmat1 support...
2026-05-26 Georgi Gerganovci : remove vulkan SDK dep from webgpu job (#23718)
2026-05-26 Max Krasnyanskyhexagon: add support for CONCAT op (#23648)
2026-05-26 Georgi Gerganovci : move more CPU jobs to self-hosted runners (#23715)
2026-05-26 Georgi Gerganovci : move sanitizer jobs to self-hosted runners (#23713)
2026-05-26 Georgi Gerganovci : reduce (disable SYCL and CANN builds/releases...
2026-05-26 ghlegconvert : support Gemma4ForCausalLM architecture (...
2026-05-26 Michael Wandmodels : Attach Mistral3 NVFP4 weight scales (#23629)
2026-05-26 Alexey KopytkoSYCL: implement ggml_sycl_pool_vmm (#22862)
2026-05-26 Jeff Bolztests: test-backend-ops -j <N> to run tests in parallel...
2026-05-26 Niklas Shethmodel : add support for talkie-1930-13b (#22596)
2026-05-26 Masashi Yoshimuraggml-webgpu: Add MMVQ path for Q4/Q8/Q2_K/Q4_K and...
2026-05-26 Nikhil Jain[WebGPU] Check batch_compute_passes before sending...
2026-05-26 Johannes GäßlerCUDA: missing PDL sync for FWHT, better fallback (...
2026-05-25 forforever73metal : add apple device id (#23566)
2026-05-25 Max Krasnyanskysnapdragon: bump toolchain docker to v0.7 to fix ui...
2026-05-25 Georgi Gerganovci : reduce PR jobs by matching backend paths (#23675)
2026-05-25 Pascalmodel: tag ffn_latent as MUL_MAT to fix buft probe...
2026-05-25 Aman GuptaCUDA: add fast walsh-hadamard transform (#23615)
2026-05-25 Pascalui: fix stop/continue during an agentic loop (#23356)
2026-05-25 Michael Wandconvert : add compressed-tensors NVFP4 support (#21095)
2026-05-25 Georgi Gerganovsync : ggml
2026-05-25 Georgi Gerganovggml : bump version to 0.13.0 (ggml/1510)
2026-05-25 Georgi Gerganovsync : ggml
2026-05-25 Georgi Gerganovggml : bump version to 0.12.1 (ggml/1508)
2026-05-25 Ori Pekelmanggml.h: correct ggml_silu_back arg docstring (a=dy...
2026-05-25 Dev-X25874ggml-alloc: fix out-of-bounds read in ggml_dyn_tallocr_...
2026-05-25 Johannes GäßlerTP: fix ggml context size calculation (#22616)
2026-05-25 Gilad S.ggml: `gguf_init_from_callback` and `gguf_init_from_buf...
2026-05-25 Aman Guptaserver: MTP layer kv-cache should respect draft type...
2026-05-25 alex-spacemitci : update spacemit toolchain url and enhance curl...
2026-05-25 Sigbjørn Skjæretci : fix pre-tokenizer-hashes check (#23651)
2026-05-25 Tim Neumannllama : document that only one on-device state can...
2026-05-25 Aldehir Rojasci : install host compiler on android-ndk build (#23630)
2026-05-25 Jeff Bolzggml : Parallelize quant LUT init (#23595)
next