]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2026-07-27 BeinseziiDisable -ffast-math on HIP (#25495)
2026-07-27 Xuan-Son Nguyenmtmd: support MiMo-V2.5 audio input (RVQ-based model...
2026-07-27 Adrien Gallouëtcommon : add common_print_available_devices() (#26170)
2026-07-27 zqlmodel: Add support for Nanbeige4.2 (#25994)
2026-07-27 Jonas Jankaitisfit : count nextn (MTP) blocks in n_gpu_layers so front...
2026-07-27 Titaniumtownsycl(build): parallelize ocloc invocations (#25903)
2026-07-27 Georgi Gerganovggml : adjust logic for offloading ops to weight's...
2026-07-27 Georgi Gerganovtests : remove unnecessary sync in test-save-load-state...
2026-07-27 Pascalcommon: fix explicit -md precedence over draft sidecar...
2026-07-27 Xuan-Son Nguyendocs: add exception about weight folding (#26168)
2026-07-27 shalinib-ibmggml-cpu: Enable BF16 tiled gemm optimization on PowerP...
2026-07-27 Aaron Teoargs: add `-lm mlock` where it mlocks but doesnt mmap...
2026-07-27 Pascalserver + ui: fix stream routes for model names containi...
2026-07-27 rankaiyxui: Fix symbolic math tool JS sandbox prompt (#26131)
2026-07-26 timkhronosmtmd: Add Vision Support for Minimax-M3 (#25113)
2026-07-26 Xuan-Son Nguyenmtmd: fix android build (#26150)
2026-07-26 Piero Evangelistaui: fix system message edit box not expanding to fit...
2026-07-26 BartowskiKeep Minimax's indexer tensors at F32 for speed and...
2026-07-26 Pascalui: detect the conversation import format from file...
2026-07-26 Xuan-Son Nguyencommon: add `subproc.h` wrapper, disabled on android...
2026-07-26 Eric Hartfordmtmd: add GLM-5.2-Vision (#26126)
2026-07-26 timkhronosmodel: Add MiniMax-M3 (MSA: MiniMax Sparse Attention...
2026-07-26 yzyyzyhhhopencl: fix fused RMS norm mul view offset (#26085)
2026-07-26 Pascalui: fix context gauge card regressions and land at...
2026-07-26 Reese Levineggml-webgpu: Fix WASM compilation with OpenMP (#25943)
2026-07-25 Nicky Mouhacommon : use-after-free when loading LoRA adapter fails...
2026-07-25 Xuan-Son Nguyenserver: support MCP stdio (#26062)
2026-07-25 Todor Boinovskihexagon: partial im2col support (#26007)
2026-07-25 hogeheer499... common : skip empty implicit default preset (#25643)
2026-07-25 Xuan-Son Nguyenserver: add format arg to datetime tool (#26117)
2026-07-25 Tekin Ertekinserver : add missing task parameters(adaptive_target...
2026-07-25 Adrien Gallouëtvendor : update cpp-httplib to 0.51.0 (#26067)
2026-07-25 Yongmin Yoo 유용민Update ggml/src/gguf.cpp : Defined virtual keyword...
2026-07-25 Aldehir Rojascommon : add support for multiple end sequences in...
2026-07-25 helanfxztests: synchronize save-load-state generation (#26056)
2026-07-24 Zach Winterui: reduce per-token render cost when streaming (...
2026-07-24 Pascalui: remove render effects (#26083)
2026-07-24 Pedro Cuencamodel: add GLM 5.2 Indexer support (#25407)
2026-07-24 Pascalui: fix MCP server display name conflicts in tools...
2026-07-24 Nigel Boschserver: support "reasoning_effort": "none" in OAI API...
2026-07-24 Xuan-Son Nguyenllama: various bug fixes (#26051)
2026-07-24 Johannes GäßlerHIP: remove rocWMMA FlashAttention (#26046)
2026-07-24 Hongqiang Wangopencl: cache compiled cl_program binaries on disk...
2026-07-24 kumaalopencl: do not treat NULL-mask flash attention as causa...
2026-07-24 Xuan-Son Nguyenskill: create `add-new-model` and `code-review` (#26042)
2026-07-24 PascalUI: Fix settings precedence, Factory < Admin (--ui...
2026-07-24 Matt Thompsoncohere2 moe template parser: enforce JSON schema for...
2026-07-24 Xuan-Son Nguyenvendor: update subprocess.h (#26061)
2026-07-24 Max Krasnyanskyhexagon: further improved pipeline of the core bits...
2026-07-23 adgup-qtihexagon: fix Windows crash when op_poll is enabled...
2026-07-23 Johannes GäßlerCUDA: fix external compilation of q1_0 MMQ (#25778)
2026-07-23 Aaron Teoargs: refactor mlock/mmap/directio into load-mode ...
2026-07-23 Pascalcontrib: fix leftovers from the AI usage policy update...
2026-07-23 Ilia Ilmermetal : add f16 type support to leaky relu (#25981)
2026-07-23 Shahir BIn... conversion: fix non-MoE NomicBert GGUF conversion error...
2026-07-22 Xuan-Son Nguyencontrib: allow all AI-generated code in general (#26012)
2026-07-22 Pascalui: Add a "Default" option for the reasoning selector...
2026-07-22 Oliver SimonsCUDA: Improve NVFP4 W4A4 activation quantization (...
2026-07-22 Todor Boinovskihexagon: activation ops update (#25974)
2026-07-22 Niklas Wenzelmtmd: use RAII for setting and resetting non-causal...
2026-07-22 rankaiyxfeat(ui): add symbolic math support to JS sandbox via...
2026-07-22 Piotr Wilkin... minor: fix reasoning preserve var for DS4 [no ci] ...
2026-07-22 Pascalcommon: infer the speculative type from the draft repo...
2026-07-22 Piotr Wilkin... Fix DeepSeek4 crafted template (#25414)
2026-07-22 shalinib-ibmggml: enable PowerPC backend variants on AIX (#25983)
2026-07-22 KyleHagyci : fix SYCL package shared library lookup (#25987)
2026-07-22 m1elwebgpu : add CONV_2D_DW (depthwise conv2d) kernel ...
2026-07-22 Pascalcuda: GET_ROWS quants (#25962)
2026-07-22 helanfxzllama-arch: fix DeepSeek4 APE tensor op (#25945)
2026-07-22 Joe RowellAdd support for Laguna XS.2 & M.1 (#25165)
2026-07-21 wendadawenconvert: fix handle HunyuanVL XD-RoPE config (#25514)
2026-07-21 Gerben van Vmtmd : use align_corners for qwen3vl vision position...
2026-07-21 Wei Wanghexagon: check tensor type when reusing descriptors...
2026-07-21 Aman Guptacuda: add sqrt_softplus in topk-moe for dsv4 (#25896)
2026-07-21 Kamalesh VSkleidiai : warn once when a weight type has no KleidiAI...
2026-07-21 Pascalcommon: resolve draft repo to its requested sidecar...
2026-07-21 Pascalserver: return 400 instead of 500 on validation error...
2026-07-21 fairydreamingserver : properly handle null llama_context (#25868)
2026-07-21 Winston Mavulkan: Refactor vk_queue to use per-instance mutexes...
2026-07-21 Markus Ebnerggml-openvino: Add GGML_BACKEND_DL_IMPL invocation...
2026-07-21 Piotr Wilkin... CUDA: vectorize same-type get_rows with int4 copy ...
2026-07-20 Todor Boinovskihexagon: add CLAMP op (#25934)
2026-07-20 Aleksander... ui: Sidebar Conversations Bulk Action + Improved Settin...
2026-07-20 Aman Guptallama_dsv4: write only used rows in state (#25325)
2026-07-20 Pascalui: fix collapsed user bubble with markdown rendering...
2026-07-20 PascalUI: fix Settings/Display tool call content toggle ...
2026-07-20 Pascalui: enable the agentic flow when only the JS sandbox...
2026-07-20 Hongqiang Wangopencl: Support broadcast for Adreno MUL_MAT and honor...
2026-07-18 Ruixiang Wangmodel: rotate injected K/V cache for DFlash (#25823)
2026-07-18 Yash Raj Pandeyllama-quant : exclude i32 ffn_gate_tid2eid routing...
2026-07-17 lhezopencl: load and use `kernel_gemm_moe_q6_k_f32_ns`...
2026-07-17 Hongqiang Wangopencl: read/write MoE dp4a activation tiles to local...
2026-07-17 Hongqiang Wangopencl: transpose q4_K noshuffle scales for coalesced...
2026-07-17 Georgi Gerganovsync : ggml
2026-07-17 Georgi Gerganovggml : bump version to 0.17.0 (ggml/1568)
2026-07-17 fairydreamingtests : initialize all tensors in test_dsv4_hc to avoid...
2026-07-17 Georgi Gerganovcommon : auto-download dflash- and eagle3- HF sidecars...
2026-07-17 Aaron Teoggml-blas: default hadamard mul_mat to cpu routine...
2026-07-17 Jeff Bolzvulkan: Support Q2_0 (#25430)
2026-07-17 Todd Malsbarysycl: fix row calculation when K_QUANTS_PER_ITERATION...
next