]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2026-06-17 Georgi Gerganovmetal : implement rope_back operator (#24725)
2026-06-17 Georgi Gerganovmetal : add f16 and bf16 support for concat operator...
2026-06-17 Xuan-Son Nguyenserver: (router) add model management API (#23976)
2026-06-17 Dev-iLllama : skip main_gpu validation when no devices are...
2026-06-17 Ruixiang Wangspec: fix segfault error on long prompts for eagle3...
2026-06-17 Neo Zhang[SYCL] add dev2dev memcpy by SYCL API (#24476)
2026-06-17 Neo Zhang[SYCL] Add conv_3d (#24691)
2026-06-17 Julien Chaumondwebui: export conversations as jsonl (#24688)
2026-06-17 Winston Mavulkan: record actual memory properties during buffer...
2026-06-17 Ruben OrtlamRevert "cuda: reset cuda context after reading memory...
2026-06-17 kononnableci: fix vulkan docker images (#24595)
2026-06-17 Harapan RachmanUI : fix SSE transport detection and routing through...
2026-06-17 lhezopencl: optimize mul_mat_f16_f32_l4 for decode (#24504)
2026-06-17 Max Krasnyanskycommon: update logging to enforce max_capacity and...
2026-06-17 Zijun Yuopenvino: OV 2026.2, context-shift, Q5_1 support, gemma...
2026-06-17 Neo Zhangsycl : Enable to support fp16 by OPs: SQR, SQRT, LOG...
2026-06-17 Alexey KopytkoSYCL: fix use-after-free bug with async memcpy in MoE...
2026-06-17 Francois Dugastsycl: Add optional USM system allocations (#22526)
2026-06-16 Alessandro... vendor : update BoringSSL to 0.20260616.0 (#24693)
2026-06-16 Pascalui: add source toggle to mermaid and svg blocks (#24652)
2026-06-16 Oliver SimonsFix and restrict NVFP4 edge-cases in llama-graph (...
2026-06-16 Ruixiang Wangspec: add backend sampling support for eagle3 (#24655)
2026-06-16 Winston Mavulkan: prefer host-visible memory buffers on UMA devic...
2026-06-16 Jeff Bolzvulkan: Support gated_delta_net with S_v=16 (#24581)
2026-06-16 Ruixiang Wangspec: add spec metrics mean acceptance length and accep...
2026-06-16 Adrien Gallouëtbench : add --offline (#24511)
2026-06-16 Frosty40sycl: support reordered Q4_K/Q5_K/Q6_K MoE MUL_MAT_ID...
2026-06-16 Neo Zhang[SYCL] Support OP EXPM1, support all UT cases of FLOOR...
2026-06-16 Todd Malsbarysycl : Make GGML_SYCL_F16=ON the default (#23996)
2026-06-16 Pascalvulkan: add col2im_1d op (#24425)
2026-06-15 Tarek Dakhranchat : fix LFM2 tool-call parsing double-escaping ...
2026-06-15 Xuan-Son Nguyenmtmd: fix miscounting n_tokens (#24656)
2026-06-15 Piotr Wilkin... chat: include full unparsed prompt in debug (#24650)
2026-06-15 Julien Jerphaniondocs: Add instructions to install `llama.cpp` from...
2026-06-15 Pascalchat: harden peg-native tool call parsing (#24329)
2026-06-15 Piotr Wilkin... chat: fix an "oldie but goodie" grammar generator bug...
2026-06-15 Georgi Gerganovmtmd : add post-decode callback (#24645)
2026-06-15 Jeff Bolzvulkan: support more CONCAT types (#24579)
2026-06-15 Andreiwasm : fix fallback symbol collision (#24639)
2026-06-15 KatostrofikSYCL: use native subgroup size for K-quant DMMV (#21700)
2026-06-15 someoneinjdsycl: fix soft_max_f32 max reduction (#24451)
2026-06-15 Neo Zhangsycl : fix reorder function; add fp32/fp16 in build...
2026-06-15 Neo Zhangsycl : enhance set_rows to support q1_0, mxfp4, nvfp4...
2026-06-15 Neo Zhang[SYCL] add to support pool_1d, move pool_1d/2d code...
2026-06-15 Alexey Kopytko[SYCL]: Remove per-allocation Level Zero runtime checks...
2026-06-15 Georgi Gerganovmetal : add repeat bf16 (#24638)
2026-06-15 Piotr Wilkin... chat: fix whitespace problems once and for all (#24624)
2026-06-15 PascalUI/svg block rendering (#24080)
2026-06-15 leonardHONGCUDA: only support F32/F16 for GGML_OP_REPEAT (#24533)
2026-06-15 Masashi Yoshimuraggml-webgpu: improve i-quants mul_mat performance and...
2026-06-14 Sigbjørn Skjæretconvert : fix lora base model arch retrieval (#24621)
2026-06-14 franitelfix(ui): render thinking/reasoning block content as...
2026-06-14 Nicolas Mowenui: Add HEIC/HEIF image support (#24137)
2026-06-14 Piotr Wilkin... chat: add dedicated Cohere2MoE (North Code) parser...
2026-06-14 Mohammad Athardocs : fix typos in CUDA-FEDORA.md and grammars/README...
2026-06-14 Alexander Batischevdocker: specify registry to simplify Podman builds...
2026-06-14 PascalUI/mobile keyboard and pwa popup fixes (#24610)
2026-06-14 Amos Wongui : fix ui clipping in mobile due to incorrect height...
2026-06-14 Sigbjørn Skjæretjinja : add count/d/e filter aliases (#24606)
2026-06-14 Michael Wandcli : fix not copying preserved tokens (#24258)
2026-06-14 BartowskiAdd cohere2moe to llama-vocab for TINY_AYA (#24601)
2026-06-14 Sigbjørn Skjæretci : use CUDA label for cuda backend (#24594)
2026-06-14 Sigbjørn Skjæretadd sycl to check-release (#24583)
2026-06-13 Aldehir Rojasui : fix llama-ui-embed crash when no asset dir is...
2026-06-13 Michael WandAdd arch support for cohere2-MoE (#24260)
2026-06-13 Sigbjørn Skjæretjinja : fix negative step slice with start/stop values...
2026-06-13 Xuan-Son Nguyenui: build-time gzip compression (#24571)
2026-06-13 Sigbjørn Skjæretjinja : fix split and replace with empty first arg...
2026-06-13 Jeff Bolzvulkan: support non-contig unary/glu ops (#24215)
2026-06-13 Xuan-Son Nguyenui: keep original file name and path (#24568)
2026-06-13 Xuan-Son Nguyenserver: clean up static assets handling (#24550)
2026-06-13 Georgi Gerganovfit : wrap llama_device_memory_data (#24522)
2026-06-13 Muhammad Salemfix sycl links in release notes (#24527)
2026-06-12 Xuan-Son Nguyenmtmd: add batching API (#24384)
2026-06-12 Sigbjørn Skjæretci : unbreak release harder (#24545)
2026-06-12 Sigbjørn Skjæretci : unbreak release (#24544)
2026-06-12 Georgi Gerganovserver : fix reasoning budget WebUI precedence over...
2026-06-12 Ruben Ortlamvulkan: add pipeline barriers for memcpy read operation...
2026-06-12 Aleksander... ui: PWA support (#23871)
2026-06-12 Georgi Gerganovfit : avoid including llama-ext.h in fit.h (#24506)
2026-06-12 Georgi Gerganovsync : ggml
2026-06-12 Georgi Gerganovggml : bump version to 0.15.1 (ggml/1541)
2026-06-12 Adrien Gallouëtvendor : update cpp-httplib to 0.47.0 (#24395)
2026-06-12 PascalUI/jpeg exif orientation (#24196)
2026-06-12 Ruixiang Wangspec: add EAGLE3 speculative decoding support (#18039)
2026-06-12 ZihaoMuggml: support concat for scalar types at cuda backend...
2026-06-12 Neo Zhang[SYCL] Fix CI build & release for SYCL backend (#24387)
2026-06-12 shaofeiqiopencl: add q5_0/q5_1 gemm and gemv kernels for Adreno...
2026-06-11 wencandocker : support specifying the GCC version for CUDA...
2026-06-11 Jeff Bolzvulkan: ifdef eMesaHoneykrisp (build fix) (#24479)
2026-06-11 Georgi Gerganovsync : ggml
2026-06-11 Georgi Gerganovggml : bump version to 0.15.0 (ggml/1539)
2026-06-11 Winston Mavulkan: add fast path for contiguous buffer transfers...
2026-06-11 Kevin Liuvulkan: use medium matmul tile on Asahi Linux (#24306)
2026-06-11 Xuan-Son Nguyenserver: skip unused log lines on router mode (#24463)
2026-06-11 o7sivocab : adopt leading TemplateProcessing special token...
2026-06-11 o7sivocab : refactor normalizer flags into options struct...
2026-06-11 Aldehir Rojasserver : skip checkpoints beyond pos_next (#24411)
2026-06-10 Adrien Gallouëtvendor : update LibreSSL to 4.3.2 (#24397)
2026-06-10 Gaurav GargRemove padding and multiple D2D copies for MTP (#24086)
next