]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2026-04-03 Piotr Wilkin... fix: remove stale assert (#21369)
2026-04-03 uvosHIP: build eatch ci build test for a different architec...
2026-04-03 Tillerinofix: add openssl to nix dependencies (#21353) (#21355)
2026-04-03 Vishal Singhggml-zendnn : add MUL_MAT_ID op support for MoE models...
2026-04-03 Piotr Wilkin... vocab: fix Gemma4 tokenizer (#21343)
2026-04-03 Radoslav Gerganovrpc : reuse compute graph buffers (#21299)
2026-04-03 Georgi Gerganovchat : avoid including json in chat.h (#21306)
2026-04-03 Georgi Gerganov(revert) kv-cache : do not quantize SWA KV cache (...
2026-04-03 Vishal Singhci : add AMD ZenDNN label to PR labeler (#21345)
2026-04-02 Slobodan Josic[HIP] Bump ROCm version to 7.2.1 (#21066)
2026-04-02 Piotr Wilkin... fix: gemma 4 template (#21326)
2026-04-02 Bartowskitests : add unit test coverage for llama_tensor_get_typ...
2026-04-02 Zheyuan Chenggml-webgpu: add vectorized flash attention (#20709)
2026-04-02 Ruben Ortlamtests: allow exporting graph ops from HF file without...
2026-04-02 Xuan-Son Nguyenmodel, mtmd: fix gguf conversion for audio/vision mmpro...
2026-04-02 Aldehir Rojascommon : add commentary rules for gpt-oss-20b (#21286)
2026-04-02 Piotr Wilkin... Relax prefill parser to allow space. (#21240)
2026-04-02 Jesus Talaverachat : add Granite 4.0 chat template with correct tool_...
2026-04-02 Georgi Gerganovkv-cache : do not quantize SWA KV cache (#21277)
2026-04-02 Roger ChenIgnore Transfer-Encoding header. (#20269)
2026-04-02 Georgi Gerganovsync : ggml
2026-04-02 Georgi Gerganovggml : bump version to 0.9.11 (ggml/1456)
2026-04-02 Neo Zhangsycl : fix llama_kv_cache hang when kv_cache is huge...
2026-04-02 Todor Boinovskihexagon : add cumsum op support (#21246)
2026-04-01 Xuan-Son Nguyencontrib : rewrite AGENTS.md, make it more clear about...
2026-04-01 lhezopencl: fix leak in Adreno q8_0 path (#21212)
2026-04-01 Aleksander... server: Bypass API Key validation for WebUI static...
2026-04-01 Johannes GäßlerCUDA: fix FA kernel selection logic (#21271)
2026-04-01 Martin Klacerkleidiai: add CPU feature detection to CI run script...
2026-04-01 Nikhil JainUpdate Dawn version in WebGPU CI (#20784)
2026-04-01 Aparna M Phexagon: improve RMS_NORM and DIV accuracy (#21251)
2026-04-01 Jonathanfix: tool call parsing for LFM2 and LFM2.5 models ...
2026-04-01 Georgi Gerganovllama : rotate activations for better quantization...
2026-04-01 Xuan-Son Nguyenscripts: add function call test script (#21234)
2026-04-01 Georgi Gerganovsync : ggml
2026-04-01 Georgi Gerganovggml : bump version to 0.9.10 (ggml/1454)
2026-04-01 Neo Zhangsycl : support nvfp4 type in mul_mat (#21227)
2026-04-01 Michael Wandggml-cuda: Add generic NVFP4 MMQ kernel (#21074)
2026-04-01 Ettore Di Giacintomemory: respect unified KV cache in hybrid memory for...
2026-04-01 uvosCUDA/HIP: Fix kernel slection for mmvq mmid kernel...
2026-04-01 Georgi Gerganovggml : fix RWKV ops thread assignment (#21226) upstream/0.0.8611
2026-04-01 Taimur Ahmadggml-cpu: fix fallback for RVV kernels without zvfh...
2026-04-01 Anav PrasadCUDA: Add Flash Attention Support for Head Dimension...
2026-04-01 Ed Addariollama : refactor llama_model_quantize_params to expose...
2026-04-01 Reese Levineggml webgpu: quantized buffers to u32 + wider browser...
2026-03-31 Abhijit Rameshggml-webgpu: port all AOT operators to JIT (#20728)
2026-03-31 Aleksander... fix: Use lower-case proxy headers naming (#21235)
2026-03-31 Adrien Gallouëtcommon : cleanup logs and modernize the progress bar...
2026-03-31 hipuddingCANN: fix multi-thread set_tensor race conditions ...
2026-03-31 Xuan-Son Nguyenserver: (webui) no more gzip compression (#21073)
2026-03-31 Aldehir Rojascommon : gpt-oss handle builtin and unsolicited tool...
2026-03-31 lainon1fix: correct misspellings in code comments (#21217)
2026-03-31 Seungmin KimCI: Enable CPU and Vulkan ARM64 Release (#21207)
2026-03-31 Georgi Gerganovsync : ggml
2026-03-31 Georgi Gerganovggml : bump version to 0.9.9 (ggml/1449)
2026-03-31 Adrien Gallouëtcommon : move up common_init() and fix Windows UTF...
2026-03-31 Neo Zhangsycl : enhance fattn perf (#21185)
2026-03-31 mtmcpcommon: add bounds check in common_init_result::sampler...
2026-03-31 SATISH K Cfix: include API key in CORS proxy requests for MCP...
2026-03-31 Piotr Wilkin... server/webui: cleanup dual representation approach...
2026-03-31 Adrien Gallouëtvendor : update BoringSSL to 0.20260327.0 (#21211)
2026-03-31 Galunidcommon : Disable backend sampling if reasoning budget...
2026-03-30 shaofeiqiopencl: add q4_K gemm and gemv kernels for Adreno ...
2026-03-30 Seungmin KimCI : Enable CUDA and Vulkan ARM64 runners and fix CI...
2026-03-30 Zhihao "Zephyr... jinja : handle empty expressions correctly (#20913)
2026-03-30 Oliver SimonsCUDA : Fix CUB's argsort when nrows % block_size =...
2026-03-30 Radoslav Gerganovrpc : fix misleading error log (#21184)
2026-03-30 Aleksander... webui: Fix branching logic on edit message (#21175)
2026-03-30 Aman Guptallama-model-loader: print warning when using overrides...
2026-03-30 Sigbjørn Skjæretci : bump ty to 0.0.26 (#21156)
2026-03-30 Xuan-Son Nguyenserver: wrap headers for mcp proxy (#21072)
2026-03-29 Sigbjørn Skjæretadd missing ROPE_FACTORS_LONG/SHORT for MiniCPM (#21150)
2026-03-29 Gaurav GargOptimize MOE GEMV kernel for BS > 1. (#20905)
2026-03-29 Max Krasnyanskyhexagon: dma optimizations (mostly fixing regressions...
2026-03-29 Davi Henrique... devops: including compute-runtime for intel.Dockerfile...
2026-03-29 Neo Zhang[SYCL] Enhance build script to use half cores to build...
2026-03-28 Sigbjørn Skjæretfix **/x glob matching (#21129)
2026-03-28 Piotr Wilkin... common/parser: fix handling of tool definition with...
2026-03-28 Sigbjørn Skjæretcommon : add character class support to glob_match...
2026-03-28 BlueMöhreWebUI: Replace illegal nested button elements (#21026)
2026-03-28 Adriencommon/json-schema: fix: handle non-capturing groups...
2026-03-28 Aldehir Rojascommon : add reasoning_format = none support to gpt...
2026-03-28 Georgi Gerganovserver : fix processing of multiple back-to-back mtmd...
2026-03-28 Adrien Gallouëtci : gracefully shut down the server (#21110)
2026-03-28 Woof DogDocument custom default webui preferences in server...
2026-03-28 Aleksander... webui: Conversation forking + branching improvements...
2026-03-28 Adrien Gallouëtvendor : update cpp-httplib to 0.40.0 (#21100)
2026-03-28 Ruben Ortlamvulkan: add noncontiguous GLU support (#21081)
2026-03-28 Piotr Wilkin... common/parser: fix reasoning whitespace bugs + extra...
2026-03-28 Sigbjørn Skjæretcli : add /glob command (#21084)
2026-03-28 Ts-sounddocker : fix and enable ARM64 image build (#20929)
2026-03-28 Adrien Gallouëtserver : add custom socket options to disable SO_REUSEP...
2026-03-27 Aldehir Rojascommon : inhibit lazy grammar sampler while reasoning...
2026-03-27 Kusha Gharahiserver: Introduce LLAMA_BUILD_WEBUI build flag to allow...
2026-03-27 Yiwei Shaohexagon: support for IQ4_NL and MXFP4 (#21018)
2026-03-27 Aleksander... webui: Improve Chat Messages initial scroll + auto...
2026-03-27 AN Longserver: remove the verbose_prompt parameter (#21059)
2026-03-27 Xuan-Son Nguyenmtmd: add more sanity checks (#21047)
2026-03-27 Xuan-Son Nguyenserver: add built-in tools backend support (#20898)
2026-03-27 Radoslav Gerganovrpc : proper handling of data pointers to CPU buffers...
next