]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2024-02-21 Ettore Di Giacintoreadme : add LocalAI to the availables UI (#5629)
2024-02-21 Georgi Gerganovsync : ggml (#5633)
2024-02-21 Georgi Gerganovreadme : update hot topics
2024-02-21 Daniel Beveniusllava : add --skip-unknown to 1.6 convert.py (#5632)
2024-02-21 postmastersllama : add `gemma` model (#5631)
2024-02-21 Meng, Hengyu[SYCL] conext add name (#5624)
2024-02-21 KawrakowIQ4_NL: 4-bit non-linear quants with blocks of 32 ...
2024-02-20 CJ Paisserver : support llava 1.6 (#5553)
2024-02-20 slarenmake : fix debug build with CUDA (#5616)
2024-02-20 Daniel Beveniusllava : add explicit instructions for llava-1.6 (#5611)
2024-02-20 Xuan Son NguyenServer: use llama_chat_apply_template (#5593)
2024-02-20 Dane Madsenreadme : update UI list (#5605)
2024-02-20 Haoxiang Feimetal : add build system support for embedded metal...
2024-02-20 Pierrick Hymbertserver : health endpoint configurable failure on no...
2024-02-20 AidanBeltonSUpdate ggml_sycl_op_mul_mat_vec_q (#5502)
2024-02-19 Mathijs de... nix: now that we can do so, allow MacOS to build Vulkan...
2024-02-19 0cc4mEnable Vulkan MacOS CI
2024-02-19 0cc4mRefactor validation and enumeration platform checks...
2024-02-19 0cc4mAdd check for VK_KHR_portability_enumeration for Molten...
2024-02-19 Mathijs de... Add preprocessor checks for Apple devices.
2024-02-19 Mathijs de... Resolve ErrorIncompatibleDriver with Vulkan on MacOS.
2024-02-19 Mathijs de... Allow for Vulkan build with Accelerate.
2024-02-19 slarencuda : ignore peer access already enabled errors (...
2024-02-19 Jared Van Bortelmake : pass CPPFLAGS directly to nvcc, not via -Xcompil...
2024-02-19 nopperlexamples : support minItems/maxItems in JSON grammar...
2024-02-19 Georgi Gerganovllava : remove extra cont (#5587)
2024-02-19 slarenllava : replace ggml_cpy with ggml_cont
2024-02-19 Georgi Gerganovsync : ggml
2024-02-19 Georgi Gerganovggml-alloc : apply ggml/731
2024-02-19 Didzis Goskometal : option to embed MSL source into compiled binary...
2024-02-19 Georgi Gerganovci : enable -Werror for CUDA builds (#5579)
2024-02-19 Georgi Gerganovmake : fix CUDA build (#5580)
2024-02-19 valirayreadme : fix typo in README-sycl.md (#5353)
2024-02-19 Abhilash Majumdercmake : remove obsolete sycl compile flags (#5581)
2024-02-19 Georgi Gerganovminor : fix trailing whitespace (#5538)
2024-02-19 Daniel Beveniusllava : avoid changing the original BakLLaVA model...
2024-02-19 NawafAlansaribaby-llama : allocate graphs in ggml_context (#5573)
2024-02-19 Xuan Son Nguyenllama : add llama_chat_apply_template() (#5538)
2024-02-19 slarencuda, metal : fix nans in soft_max (#5574)
2024-02-19 Mirko185readme : update (#5572)
2024-02-19 bmwlggml : android and old glibc NUMA incompatibility bugfi...
2024-02-18 Jared Van Bortelbuild : pass all warning flags to nvcc via -Xcompiler...
2024-02-18 Georgi Gerganovggml : restore vec dot stride arg names (#5453)
2024-02-18 Georgi Gerganovci : fix wikitext url + compile warnings (#5569)
2024-02-18 Georgi Gerganovmetal : fix unused warnings (#0)
2024-02-18 Robey Holderithcommon, server : surface min_keep as its own parameter...
2024-02-18 Pierrick Hymbertserver : slots monitoring endpoint (#5550)
2024-02-18 Georgi Gerganovsampling : do not set min_keep to n_probs (#5564)
2024-02-18 Georgi Gerganovcmake : fix GGML_USE_SYCL typo (#5555)
2024-02-18 Pierrick Hymbertserver : enhanced health endpoint (#5548)
2024-02-18 Pierrick Hymbertserver : --n-predict option document and cap to max...
2024-02-18 Daniel Hiltgenserver : graceful server shutdown (#5244)
2024-02-18 Georgi Gerganovcommon : fix ub (#5530)
2024-02-18 Herman Semenovggml, common, examples, tests : fixed type arguments...
2024-02-18 Daniel Beveniusllava : update surgery script to not remove tensors...
2024-02-18 Kawrakow1.5 bit quantization (#5453)
2024-02-18 github-actions... flake.lock: Update
2024-02-17 Georgi Gerganovggml : add ALiBi support for ggml_soft_max_ext (#5488)
2024-02-17 Ananta Bastolaci : add an option to fail on compile warning (#3952)
2024-02-17 clibdevgitignore : update for CLion IDE (#5544)
2024-02-16 Georgi Gerganovcmake : fix VULKAN and ROCm builds (#5525)
2024-02-16 Georgi Gerganovscripts : add helpers script for bench comparing commit...
2024-02-16 Herman Semenovllava : removed excess free(NULL) operation (#5531)
2024-02-16 Herman Semenovllama : minor fixed return int value (#5529)
2024-02-16 Alexey Parfenovserver : add "samplers" param to control the samplers...
2024-02-16 Rőczey Barnabásserver : fix system prompt cli (#5516)
2024-02-16 bmwlggml : add numa options (#5377)
2024-02-16 Daniel Beveniusllava : fix clip-model-is-vision flag in README.md...
2024-02-16 Georgi Gerganovci : fix BERT model download and convert
2024-02-15 Douglas HanleyUse correct type of pooling for embedding models (...
2024-02-15 Georgi Gerganovclip : fix wrong loop condition
2024-02-15 slarencuda : print message when initialization fails (#5512)
2024-02-15 Georgi Gerganovscripts : add hf.sh helper script (#5501)
2024-02-15 Michaël de... fix(gguf-py): special tokens are no longer skipped...
2024-02-15 Elbiosllava : fix memory management bug (#5491)
2024-02-15 Johnllaba : hotfix for llava-1.6 image number (#5495)
2024-02-15 Neuman Vongvulkan: Find optimal memory type but with fallback...
2024-02-14 Runereadme : fix typo (#5490)
2024-02-14 Johnllava : update README.md (#5489)
2024-02-14 Michael Podvitskiycmake : ARM intrinsics detection for MSVC (#5401)
2024-02-14 Johnllava : support v1.6 (#5267)
2024-02-13 ATEarly return for zero size calls to get_tensor. (#5482)
2024-02-13 Johngguf : add python reader example (#5216)
2024-02-13 Jared Van Bortelllama : add support for Nomic Embed (#5468)
2024-02-13 Aarni Koskelallama : allow raw byte in SPM vocabs; don't crash on...
2024-02-13 Aarni Koskelallama : make load error reporting more granular (#5477)
2024-02-13 Daniel Beveniusfinetune : rename feed-forward tensors (w1/w2/w3) ...
2024-02-13 Georgi Gerganovtests : multi-thread the tokenizer tests (#5474)
2024-02-13 Douglas Hanleyllama : support batched embeddings (#5466)
2024-02-13 Johannes Gäßlermake: add error message for bad CUDA version (#5444)
2024-02-13 Georgi Gerganovbert : add tests + fix quantization (#5475)
2024-02-13 Georgi Gerganovtests : disable moe test (#5473)
2024-02-13 Kawrakowggml-quants : fix compiler warnings (shadow variable...
2024-02-12 Georgi Gerganovllama : fix quantization when tensors are missing ...
2024-02-12 Georgi Gerganovswift : package no longer use ggml dependency (#5465)
2024-02-12 Leepy : fix persimmon `n_rot` conversion (#5460)
2024-02-12 Abhilash Majumderggml-sycl: Replace 3d ops with macro (#5458)
2024-02-12 Daniel Beveniusllava : remove prog parameter from ArgumentParser ...
2024-02-12 Georgi Gerganovsync : ggml (#5452)
2024-02-11 Johannes GäßlerCUDA: mul_mat_vec_q tiling, refactor mul mat logic...
next