]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2025-03-07 vmobilisggml : ggml_compute_forward_concat() for arbitrary...
2025-03-07 Rémy Oggml-cpu: faster AVX2 variant for IQ1_M (#12216)
2025-03-07 Georgi Gerganovci : fix save-load test invocations (#12245)
2025-03-07 Sigbjørn Skjæretserver : Log original chat template parsing error ...
2025-03-07 Olivier Chafiksync: minja - support QwQ-32B (#12235)
2025-03-07 BB-fatmetal : simplify kernel arguments using a struct (...
2025-03-07 David HuangHIP: fix rocWMMA build flags under Windows (#12230)
2025-03-07 Daniel Beveniusmetal : fix default.metallib build (#12224)
2025-03-07 lhezopencl: Noncontiguous `norm`, `rms_norm`, disable ...
2025-03-06 xiaofeicmake : fix undefined reference errors for std::filesys...
2025-03-06 Lucas Moura... readme : update bindings (#12229)
2025-03-06 Johannes GäßlerCUDA: fix FA logic for PTX 7.0 and CC >= 7.5 (#12222)
2025-03-06 David HuangHIP: rocWMMA documentation and enabling in workflow...
2025-03-06 Olivier Chafikupdate function-calling.md w/ template override for...
2025-03-06 Aaron Teollava: add big-endian conversion for image encoder...
2025-03-06 uvosHIP/CUDA: set the paramerter value in maintain_cuda_gra...
2025-03-06 Han Yinandroid : fix KV cache log message condition (#12212)
2025-03-06 Henry Linjamäkiopencl : fix buffer alignment (#12197)
2025-03-06 Henry Linjamäkiopencl : fix `ulong` kernel args were set from `int...
2025-03-06 simon886212opencl : fix profile-related errors (#12095)
2025-03-06 Rémy Oggml-cpu: Faster IQ1 mul_mat_vec on AVX2 using BMI2...
2025-03-05 Akarshan BiswasSYCL: Disable f16 Unary OPs as not supported by the...
2025-03-05 Plamen Minevggml : fix GGMLMetalClass ODR (#12200)
2025-03-05 Daniel Beveniusci : add fetch-depth to xcframework upload (#12195)
2025-03-05 Olivier Chafik`tool-call`: fix Qwen 2.5 Coder support, add micro...
2025-03-05 Daniel Beveniusci : fix xcframework artifact tag (#12191)
2025-03-05 Daniel Beveniusci : remove xframework upload (#12190)
2025-03-05 Clauszyserver : fix cache reuse logic (#12161)
2025-03-05 Daniel Beveniusllama : add xcframework build script (#11996)
2025-03-04 mgroeber9110ggml : portability fixes for VS 2017 (#12150)
2025-03-04 Georgi Gerganovreadme : fix roadmap link (#12185)
2025-03-04 Sigbjørn Skjæretmain: allow preloading conversation with -p and add...
2025-03-04 Olivier Chafik`server`: fix deadly typo in response_format.json_schem...
2025-03-03 David HuangHIP: implement FlashAttention via rocWMMA for CDNA...
2025-03-03 Georgi Gerganovsync : ggml
2025-03-03 cmdr2cuda: unary ops as float + de-duplicate (ggml/1130)
2025-03-03 Georgi Gerganovsync : ggml
2025-03-03 cmdr2cuda/vulkan: specify fp32-only support for some operati...
2025-03-03 Georgi Gerganovsync : ggml
2025-03-03 cmdr2cuda/cpu: Increase support for fp16 unary operations...
2025-03-03 Diego Devesawhisper : support GGML_BACKEND_DL (whisper/2843)
2025-03-03 midnightcmake : fix compile assumptions for power9/etc (whisper...
2025-03-03 petterreinholdtsenTold cmake to install ggml-cpp.h as a public header...
2025-03-03 cmdr2Support pure float16 add/sub/mul/div operations in...
2025-03-03 Georgi Gerganovscripts : sync-ggml-am.sh fix
2025-03-03 Daniel Beveniusci : set GITHUB_ACTION env var for server tests (#12162)
2025-03-03 dm4tts: add speaker file support (#12048)
2025-03-03 Diego Devesatest-backend-ops : add option -p to filter by op params...
2025-03-03 ag2s20150909ggml : fix kleidiai build (#12159)
2025-03-03 Eric CurtinAdding UTF-8 support to llama.cpp (#12111)
2025-03-03 Xuan-Son Nguyenwebui : add ?m=... and ?q=... params (#12148)
2025-03-03 Akarshan BiswasSYCL: Move CPY kernels to a separate file and add few...
2025-03-02 Diego Devesaggml-backend : keep paths in native string type when...
2025-03-02 Sigbjørn Skjæretmain: use jinja chat template system prompt by default...
2025-03-01 Sigbjørn Skjæretmain: update outdated system prompt message (followup...
2025-03-01 Sigbjørn Skjæretcommon : add --system-prompt parameter, replace behavio...
2025-03-01 Erik ScholzCUDA: compress mode option and default to size (#12029)
2025-03-01 Vivianwebui : minor typo fixes (#12116)
2025-02-28 Xuan-Son Nguyenconvert : fix Norway problem when parsing YAML (#12114)
2025-02-28 William Tambelliniggml : upgrade init_tensor API to return a ggml_status...
2025-02-28 Xuan-Son Nguyenllama : add Phi-4-mini support (supersede #12099) ...
2025-02-28 Alex BrooksUpdate granite vision docs for 3.2 model (#12105)
2025-02-28 Rémy Ovulkan: add specific MMV kernels for IQ2 and IQ3 quants...
2025-02-28 Johannes GäßlerCUDA: fix logic for V100 + GGML_CUDA_FORCE_MMQ (#12098)
2025-02-28 Prashant Vithuleggml: aarch64: implement SVE kernels for q2_k_q8_k...
2025-02-28 hipuddingCANN: Fix build error with GCC 13 (#11990)
2025-02-28 Evevulkan: matmul dequantization improvements (#12015)
2025-02-28 Danielevulkan: improve im2col (#11826)
2025-02-27 Vladimir Vuksanoviccmake: Fix ggml backend dependencies and installation...
2025-02-26 Ting Loullava : add struct for FFI bindgen (#12079)
2025-02-26 Sigbjørn SkjæretRefactor gguf scripts to improve metadata handling... gguf-v0.16.0
2025-02-26 Aleksei Nikiforovgguf-py: enable reading non-native endian files (#12081)
2025-02-26 Kante Yinreadme : update infra list (#9096)
2025-02-25 Olivier Chafikdocs: add docs/function-calling.md to lighten server...
2025-02-25 Jeff Bolzvulkan: fix assertion when qy_needs_dequant (#12068)
2025-02-25 rhjdvsgsgksserver: handle echo=false on /v1/completions (#12060)
2025-02-25 Juddadd OP sigmoid (#12056)
2025-02-25 Molly Sophiaggml-cpu: Fix build with sve (#12059)
2025-02-25 Rémy Ovulkan: implement more backpropagation operators (...
2025-02-25 Olivier Chafikserver: support add_generation_prompt query param ...
2025-02-25 Alex BrooksAdd Doc for Converting Granite Vision -> GGUF (#12006)
2025-02-25 Vitali Lovichllama : expose llama_model_n_head_kv in the API (#11997)
2025-02-25 Gian-Carlo... metal : copy kernels for quant to F32/F16 conversions...
2025-02-24 lhezopencl: fix for small models (#11950)
2025-02-24 Alex Brooksllava : Add Granite Vision Support (#11794)
2025-02-24 Neo Zhang Jianyu[SYCL] Optimize mul_mat for Q4_0 on Intel GPU (#12035)
2025-02-24 Aleksei Nikiforovgguf_convert_endian.py: implement byteswapping for...
2025-02-24 Akarshan BiswasSYCL: Fix GGML_SYCL_DEBUG macro (#11995)
2025-02-23 Florent BENOITrun: allow to customize prompt by env var LLAMA_PROMPT_...
2025-02-23 Eric CurtinSome llama-run cleanups (#11973)
2025-02-22 Aaron Teoggml-cpu: Support s390x SIMD Instruction Set (#12019)
2025-02-22 Johannes GäßlerCUDA: app option to compile without FlashAttention...
2025-02-22 Ting Loullava: build clip image from pixels (#11999)
2025-02-22 Georgi Gerganovci : fix arm upload artifacts (#12024)
2025-02-22 Johannes GäßlerCUDA: optimize FA for GQA + large batches (#12014)
2025-02-22 Rohanjames1997ci : Build on Github-hosted arm64 runners (#12009)
2025-02-22 Georgi Gerganovserver : disable Nagle's algorithm (#12020)
2025-02-22 Gian-Carlo... cuda: Add Q5_1, Q5_0, Q4_1 and Q4_0 to F32 conversion...
2025-02-22 Daniel Beveniusllama.swiftui : add "Done" dismiss button to help view...
2025-02-21 Georgi Gerganovllama : skip loading unused tensors (#12004)
next