]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2024-03-14 Georgi Gerganovllama : fix typo
2024-03-14 Michael Podvitskiyllama : optimize defrag moves + fix fragmentation calcu...
2024-03-14 Ondřej Čertíkgguf-py : add support for I8, I16 and I32 (#6045)
2024-03-14 Georgi Gerganovggml : designate enum vals for integer types (#6050)
2024-03-14 Georgi Gerganovembedding : print all resulting embeddings (#899)
2024-03-14 Georgi Gerganovmetal : build metallib + fix embed path (#6015)
2024-03-14 Georgi Gerganovembedding : print cosine similarity (#899)
2024-03-13 Linwei Wangreadme : update details about running llama in Termux...
2024-03-13 Georgi Gerganovreadme : update API changes and hot topics
2024-03-13 Clint Herrongrammar : handle missing "root" node (#6004)
2024-03-13 slarenllama : add pipeline parallelism support (#6017)
2024-03-13 slarentest-backend-ops : skip CPU backend by default (#6028)
2024-03-13 AidanBeltonSUpdate get version (#6025)
2024-03-13 Xuan Son NguyenServer: Use multi-task for embeddings endpoint (#6001)
2024-03-12 slarenci : remove tidy-review (#6021)
2024-03-12 Georgi Gerganovggml : reuse quantum structs across backends (#5943)
2024-03-12 Georgi Gerganovggml : fix UB in IQ2_S and IQ3_S (#6012)
2024-03-12 Georgi Gerganovsycl : update IQ1_S kernels (WIP - not working!) (...
2024-03-11 glipticgrammar : fix unnecessarily retained pointer to rules...
2024-03-11 Kawrakow1.5 bit: we can do even better (#5999)
2024-03-11 Georgi Gerganovllama : more consistent names of count variables (...
2024-03-11 Georgi Gerganovllama : refactor unicode stuff (#5992)
2024-03-11 Jakub NUpdate server docker image URLs (#5997)
2024-03-11 Xuan Son NguyenServer: format error to json (#5961)
2024-03-11 Michael Podvitskiyggml, ci : Windows ARM runner and build fixes (#5979)
2024-03-11 Minsoo Cheongserver : maintain chat completion id for streaming...
2024-03-11 Gilad Scmake : fix subdir for `LLAMA_METAL_EMBED_LIBRARY`...
2024-03-11 Georgi Gerganovllama : fix F16/F32 downcast + improve names (#5980)
2024-03-11 KawrakowBetter 1.5 bit quantization (#5971)
2024-03-11 Abhilash Majumder[SYCL] Add q3_s and q1_s (#5886)
2024-03-11 AidanBeltonS[SYCL] Add support for SYCL Nvidia target (#5738)
2024-03-10 Georgi Gerganovmetal : move mm_id indices to shared mem (#5982)
2024-03-10 Deanandroid : fix utf8 decoding error (#5935)
2024-03-10 Georgi Gerganovreadme : update hot topics
2024-03-10 Georgi Gerganovsync : ggml
2024-03-10 Georgi Gerganovggml : try fix 32-bit arm compat (whisper/1938)
2024-03-10 Georgi Gerganovggml : remove __constant__ specifier for CUDA tables...
2024-03-10 Pierrick Hymbertserver: ci: windows build and tests (#5968)
2024-03-10 DAN™llama : add support for GritLM (#5959)
2024-03-10 Clint Herrongrammar : verify parsed state (#5950)
2024-03-10 Georgi Gerganovnix: update flake.lock (#5969)
2024-03-09 Pierrick Hymbertserver: benchmark: chat/completions scenario and other...
2024-03-09 Georgi Gerganovserver : print chat template info
2024-03-09 slarenperplexity : support using multiple sequences to allow...
2024-03-09 Georgi Gerganovreadme : update hot topics
2024-03-09 Georgi Gerganovggml : fix unnecessary f32 -> f16 -> f32 casts (mmla...
2024-03-09 Georgi Gerganovserver : fix metrics init (#5964)
2024-03-09 Georgi Gerganovggml : remove old quantization functions (#5942)
2024-03-09 Georgi Gerganovserver : clarify some items in the readme (#5957)
2024-03-09 SeungWon Jeongserver : normalize embeddings (#5956)
2024-03-09 Georgi Gerganovtests : gitignore ggml-common.h
2024-03-09 Alexey Parfenovserver : fix passing prompt as tokens (#5955)
2024-03-09 Georgi Gerganovggml : add ggml-common.h to deduplicate shared code...
2024-03-09 Georgi Gerganovserver : simplify logic for empty prompts (#5953)
2024-03-09 Xuan Son NguyenServer: reorganize some http logic (#5939)
2024-03-09 Gabe Goodhartserver : add SSL support (#5926)
2024-03-09 Pierrick Hymbertserver: tests: add truncated prompt tests, better kv...
2024-03-08 compiladellama : support Mamba Selective State Space Models...
2024-03-08 compiladellama : fix quantization of shared token_embd (#5944)
2024-03-08 Pierrick Hymbertserver: metrics: add llamacpp:prompt_seconds_total...
2024-03-08 Don Mahurinllama : assume tied weights if lm_head/output weights...
2024-03-08 Georgi Gerganovserver : fix EOS token detection with disabled cache...
2024-03-08 UEXTM.comlog : fix MSVC compile errors (#5643)
2024-03-07 Georgi Gerganovllama-bench : add embeddings option (#5924)
2024-03-07 Neo Zhang JianyuRevert "[SYCL] fix error when set main gpu to non-zero...
2024-03-07 Minsoo Cheongserver : add `/v1/completions` endpoint (#5914)
2024-03-07 Georgi Gerganovserver : refactor (#5882)
2024-03-07 Neo Zhang Jianyu[SYCL] fix error when set main gpu to non-zero (#5901)
2024-03-06 Jared Van Bortelggml : use SYS_get_cpu if SYS_getcpu is not defined...
2024-03-06 bobqianicggml : use `uint8x16_t` return type for `ggml_vqtbl1q_u...
2024-03-06 Georgi Gerganovconvert : remove AWQ remnants (#5768)
2024-03-06 Neo Zhang Jianyuadd wait() to make code stable (#5895)
2024-03-05 slarencompare-llama-bench.py : remove mul_mat_q (#5892)
2024-03-05 Jared Van Bortelquants : use MM256_SET_M128I consistently to fix gcc...
2024-03-05 ExtReMLapingrammars : blacklists character control set (#5888)
2024-03-05 Georgi GerganovRevert "grammars : don't allow to output unescaped...
2024-03-05 ExtReMLapingrammars : don't allow to output unescaped new line...
2024-03-05 0cc4mVulkan Improvements (#5835)
2024-03-05 Neo Zhang Jianyu[SYCL] fix mul_mat fault in CI/unit-test (#5862)
2024-03-05 Minsoo Cheongfix editorconfig check break (#5879)
2024-03-05 Jeffrey Quesnellefix speculative decoding build on windows (#5874)
2024-03-05 hutlinix: static build (#5814)
2024-03-04 Georgi Gerganovllama : fix embeddings (#5796)
2024-03-04 Georgi Gerganovflake : fix
2024-03-04 Georgi Gerganovggml : fix unknown status (#0)
2024-03-04 Georgi Gerganovsync : ggml
2024-03-04 Michael Podvitskiyggml : introduce ggml_status (ggml/750)
2024-03-04 Dane Madsencmake : handle cases where git index is not found in...
2024-03-04 Minsoo Cheongspeculative : implement stochastic speculative sampling...
2024-03-04 Xuan Son Nguyenadd alias for chat template (#5858)
2024-03-04 Georgi Gerganovsync : ggml
2024-03-04 leejetadd some new ops, fix some operators and add batch...
2024-03-04 DAN™common : use LLAMA_DEFAULT_SEED (#5855)
2024-03-04 DAN™main : support special tokens as reverse/anti prompt...
2024-03-03 slarencuda : fix data race in soft max (#5853)
2024-03-03 Georgi Gerganovreadme : add API changes section
2024-03-03 Douglas Hanleyllama : allow for user specified embedding pooling...
2024-03-03 Nindalethgguf-dump : support i-quants (#5841)
2024-03-03 compiladellama : fix llama_copy_state_data with fragmented KV...
2024-03-03 Pierrick Hymbertci : schedule slow server tests only on Release or...
next