]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2025-08-23 Johannes GäßlerCUDA: fix half2 -> half conversion for HIP (#15529)
2025-08-23 Jeff Bolzvulkan: optimize rms_norm, and allow the work to spread...
2025-08-23 Piotr Wilkin... model : add support for Seed-OSS (#15490)
2025-08-23 Johannes Gäßlerscripts: fix compare-llama-bench.py (#15521)
2025-08-23 LaffeyNyaachat : fix debug build assertion in trim function ...
2025-08-23 Jeff Bolzvulkan: Rewrite synchronization to allow some overlap...
2025-08-23 R0CKSTARvulkan.Dockerfile: install vulkan SDK using tarball...
2025-08-23 Aclyvulkan : support ggml_mean (#15393)
2025-08-23 Jeff Bolzvulkan: optimize mul_mat_id loading row ids into shared...
2025-08-22 Johannes Gäßlertest-opt: allow slight inprecision (#15503)
2025-08-22 Reese Levineggml WebGPU: add support for quantization types (#15440)
2025-08-22 Aldehir Rojasmodel : gpt-oss add response_format support (#15494)
2025-08-22 rmatifggml: add `conv3d` op (#15182)
2025-08-22 Yavor Ivanovcuda : add Pad Reflect 1D support (#14659)
2025-08-22 Georgi Gerganovllama : remove KV cache defragmentation logic (#15473)
2025-08-22 Aaron Teoggml-cpu: Support Q5_0 and Q5_1 on s390x (#15486)
2025-08-22 65aserver : Support multimodal completion and embeddings...
2025-08-22 Tarek Dakhranreadme : model : mtdm : lfm2 improvements (#15476)
2025-08-22 Chenguang LiCANN: Optimize RMS_NORM using cache (#15419)
2025-08-21 Diego Devesasched : fix possible use of wrong ids tensor when offlo...
2025-08-21 Georgi Gerganovllama : remove deprecated llama_kv_self API (#15472)
2025-08-21 Georgi Gerganovgraph : remove build_attn_with_sinks overload (#15469)
2025-08-21 Aclyvulkan : support conv_2d_dw with f16 weights (#15392)
2025-08-21 Dong Won Kimvulkan: add exp operation (#15456)
2025-08-21 Jeff Bolzvulkan: Reuse conversion results in prealloc_y (#15410)
2025-08-21 Jie Fu (傅杰)examples : fix some typos in examples/model-conversion...
2025-08-21 Georgi Gerganovkv-cache : drop the "unified" prefix (#15467)
2025-08-21 Jie Fu (傅杰)examples : install torch-cpu for model conversion tool...
2025-08-21 Ali Tariqci : enable RVV1.0 native build (#15386)
2025-08-21 Georgi Gerganovci : continue file download with wget (#15471)
2025-08-21 Daniel Beveniusexamples : add model conversion tool/example (#15455)
2025-08-21 Michael Gibaci : fix -Werror=return-type in clip.cpp so ci/run...
2025-08-21 Copilotci : add copilot-instructions.md (#15286)
2025-08-21 Julien Denizeconvert : make Mistral community chat templates optiona...
2025-08-21 Jie Fu (傅杰)common : fix incorrect print of non-ascii characters...
2025-08-21 Xuan-Son Nguyenggml : fix condition of im2col on Metal backend (#15460)
2025-08-21 stduhpfserver : fix webui (#15462)
2025-08-21 Daniel Beveniusexamples : remove references to `make` in examples...
2025-08-21 R0CKSTARmusa: add GGML_UNUSED_VARS (#15446)
2025-08-20 Diego Devesasched : copy only the used experts when offloading...
2025-08-20 teoserver: fix OpenAI API compatibility for usage statisti...
2025-08-20 Johannes GäßlerCUDA: refactor FA support/selection code (#15454)
2025-08-20 Johannes GäßlerCUDA: replace GGML_CUDA_F16 with CUDA arch checks ...
2025-08-20 Jeff Bolzvulkan: shorten pipeline name strings (#15431)
2025-08-20 Daniel Beveniuschat: handle gpt-oss return/end token inconsistency...
2025-08-20 Jie Fu (傅杰)common : fix context shift help message (#15448)
2025-08-20 xiaobing318cmake : fix target include directories (#15450)
2025-08-20 Daniel Beveniusmake : remove make in favor of CMake (#15449)
2025-08-20 Georgi Gerganovlookahead : add sample command to readme (#15447)
2025-08-20 R0CKSTARmusa: fix build warnings (#15258)
2025-08-19 lhezopencl: mark `argsort` unsupported if cols exceed workg...
2025-08-19 Georgi Gerganovmodel : add gpt-oss type strings (#15424)
2025-08-19 Gian-Carlo... common : Add top-nsigma sampler to help globally (...
2025-08-19 Georgi Gerganovserver : disable context shift by default (#15416)
2025-08-19 SHUAI YANGCANN: optimize rope operator (#15335)
2025-08-19 R0CKSTARmusa: handle __hgt2_mask, available starting from MUSA...
2025-08-19 Marvin Gießingggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le...
2025-08-19 Xuan-Son Nguyenchat : clarify the meaning of reasoning_format (#15408)
2025-08-19 Georgi Gerganovserver : remove swa_full warning (#15399) upstream/latest
2025-08-19 Georgi Gerganovbatched-bench : use rand tokens (#15398)
2025-08-18 Xuan-Son Nguyenmtmd : clean up clip_n_output_tokens (#15391) upstream/0.0.6199
2025-08-18 Georgi Gerganovcodeowners : remove mmv.*
2025-08-18 Georgi Gerganovsync : ggml
2025-08-18 Georgi Gerganovscripts : update sync scripts
2025-08-18 Sigbjørn Skjæretllama : merge conts and reshapes and remove unnecessary...
2025-08-18 Georgi Gerganovreadme : update hot topics (#15397)
2025-08-18 davidefserver : fix incoming tasks not process in order (...
2025-08-18 Dobri DanchevFix broken build: require updated pip to support -...
2025-08-18 compiladeggml-quants : fix make_qp_quants NANs and IQ1 assertion...
2025-08-18 Jeff Bolzvulkan: disable spirv-opt for bfloat16 shaders (#15352)
2025-08-17 Oleksandr Kuvshynovserver : export max observed n_past value (#15361)
2025-08-17 Jeff Bolzvulkan: Use larger workgroups for mul_mat_vec when...
2025-08-17 Dong Won Kimvulkan: support sqrt (#15370)
2025-08-17 Sigbjørn Skjæretconvert : force patch_embd weights to F16 or F32 to...
2025-08-17 Sigbjørn Skjæretci : fix hang in windows-hip build/release (#15365)
2025-08-17 Jeff Bolzvulkan: Optimize argsort (#15354)
2025-08-16 Tarek Dakhranmodel : support vision LiquidAI LFM2-VL family (#15347)
2025-08-16 Jeff Bolzvulkan: fuse adds (#15252)
2025-08-16 Jeff Bolzvulkan: Support mul_mat_id with f32 accumulators (...
2025-08-16 Jeff Bolzvulkan: Add missing bounds checking to scalar/coopmat1...
2025-08-16 rmatifOpenCL: add initial FA support (#14987)
2025-08-15 Daniel Beveniuscommon : fix double bos, use common_chat_templates...
2025-08-15 lhezopencl: add initial mxfp4 support via mv (#15270)
2025-08-15 Georgi Gerganovvulkan : fix out-of-bounds access in argmax kernel...
2025-08-15 Georgi Gerganovvulkan : fix compile warnings on macos (#15340)
2025-08-15 Aaron Teoggml: initial IBM zDNN backend (#14975)
2025-08-15 Sigbjørn Skjæretci : fix ios-xcode-build (#15324)
2025-08-15 Diego Devesaci : move ccache action to ggml-org fork (#15328)
2025-08-15 Johannes Gäßlertest-opt: fix backend support check (#15317)
2025-08-14 Johannes GäßlerCUDA: fix negative KV_max values in FA (#15321)
2025-08-14 Georgi Gerganoveval-callback : stop on first NaN (#15320)
2025-08-14 Diego Devesachat : include kwargs in template example (#15309)
2025-08-14 Daniel Beveniusllama : add 18-layer model type for Gemma 3-270m (...
2025-08-14 simevodevops : fix compile bug when the BASE_CUDA_DEV_CONTAIN...
2025-08-14 uvosHIP: Cleanup hipification header (#15285)
2025-08-14 Aldehir Rojasgpt-oss: implement harmony parsing (#15181) upstream/0.0.6164
2025-08-14 Christian Kastnerdocker : Enable GGML_CPU_ALL_VARIANTS for ARM (#15267)
2025-08-14 Georgi Gerganovreadme : update hot topics (#15315)
2025-08-14 Jeff Bolzvulkan: perf_logger improvements (#15246)
2025-08-14 Georgi Gerganovserver : add SWA checkpoints (#15293)
next