]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2023-04-23 slarenFix LoRA acronym (#1145)
2023-04-23 Georgi Gerganovscripts : add helper scripts to synch ggml repo
2023-04-23 DannyDaemonicAdded README.md for main with examples and explanations...
2023-04-23 Georgi Gerganovggml : do not print perf ops that have not been used...
2023-04-23 Georgi Gerganovggml : better PERF prints + support "LLAMA_PERF=1 make"
2023-04-23 Stephan WalterImprove AVX2 for vec_dot_q4_3_q8_0 (#1138)
2023-04-23 Pavol Rusnakreadme : update gpt4all instructions (#980)
2023-04-23 Yishuo WangA better `packNibbles` and `mul_sum_i8_pairs_float...
2023-04-22 Georgi Gerganovggml : fix Q4_3 cuBLAS
2023-04-22 Stephan Walterci : trigger CI for drafts, but not most PR actions...
2023-04-22 Stephan WalterFix CI: ARM NEON, quantization unit tests, editorconfig...
2023-04-22 unboundedggml : unit test for quantization functions (#953)
2023-04-22 wbpxre150llama : print timings on ctrl+c exit (#1021)
2023-04-22 eieryllama : have n_batch default to 512 (#1091)
2023-04-22 Howard Sucmake : fix build under Windows when enable BUILD_SHARE...
2023-04-22 Georgi Gerganovggml : fix AVX build + update to new Q8_0 format
2023-04-22 Georgi Gerganovggml : alternative Q4_3 implementation using modified...
2023-04-22 Stephan Walterggml : AVX2 optimization for vec_dot_q4_3_q8_0 and...
2023-04-22 Clint Herronexamples : Improve Alpaca Default Repeat Penalty: Bette...
2023-04-22 xaedesllama : add api for getting/setting the complete state...
2023-04-21 slarenImprove cuBLAS performance by using a memory pool ...
2023-04-21 apazllama : fixed rlimit error message (#888)
2023-04-21 源文雨cmake : link threads publicly to ggml (#1042)
2023-04-21 Alex Klinkhamermain : evaluate tokens in batches after swapping contex...
2023-04-21 xaedesllama : remember and restore kv cache data pointers...
2023-04-21 Kawrakowggml : a faster version for Q4_1 x Q8_0 dot products...
2023-04-21 slarenShow perplexity ETA in hours and minutes (#1096)
2023-04-21 Georgi Gerganovllama : fix comment for "output.weight" tensor
2023-04-20 Stephan WalterAdd ggml-model-*.bin checksums for 7B, 13B, 30B, 65B...
2023-04-20 Georgi Gerganovggml : sync ggml (add GPT-NeoX RoPE implementation)
2023-04-20 Georgi Gerganovggml : fix bug in ggml_compute_forward_dup_f32()
2023-04-20 slarenAdd Q4_3 support to cuBLAS (#1086)
2023-04-20 Georgi Gerganovggml : do not break cuBLAS build (Q4_3 is not yet imple...
2023-04-20 Georgi Gerganovggml : fix Q4_3 quantization
2023-04-20 Kawrakowllama : multi-threaded quantization (#1075)
2023-04-20 Georgi Gerganovggml : add Q4_3 quantization (#1082)
2023-04-20 Ivan Komarovci : remove the LLAMA_ACCELERATE matrix dimension from...
2023-04-20 源文雨fix: LLAMA_CUBLAS=1 undefined reference 'shm_open'...
2023-04-20 Stephan WalterAVX2 optimization for vec_dot_q4_2_q8_0 (#1068)
2023-04-20 slarenImprove cuBLAS performance by dequantizing on the GPU...
2023-04-19 CRD716Minor: Readme fixed grammar, spelling, and misc updates...
2023-04-19 KawrakowQ4_2 quantization with rmse-optimized scale and quants...
2023-04-19 Georgi Gerganovggml : use 8-bit precision for Q4_1 intermediate result...
2023-04-19 Georgi Gerganovreadme : add warning about Q4_2 and Q4_3
2023-04-19 Stephan Walterggml : Q4 cleanup - remove 4-bit dot product code ...
2023-04-19 slarenAdd NVIDIA cuBLAS support (#1044)
2023-04-18 slarenMulti-threaded ggml_cpy (#1035)
2023-04-18 Georgi Gerganovggml : add new Q4_2 quantization (ARM only) (#1046)
2023-04-18 Georgi Gerganovggml : scratch that - vmlaq_n_f32 is always better
2023-04-18 Georgi Gerganovgitignore : vdot
2023-04-18 Georgi Gerganovggml : optimize ggml_vec_dot_q4_0_q8_0() using vectoriz...
2023-04-18 KawrakowAdding a simple program to measure speed of dot product...
2023-04-18 Georgi Gerganovreadme : update hot topics about new LoRA functionality
2023-04-18 Georgi Gerganovci : do not run on drafts
2023-04-18 Ivan KomarovDo not close file after mmap (Windows version) (#1034)
2023-04-17 Atsushi Tatsumareadme : add Ruby bindings (#1029)
2023-04-17 Cameronadd 4_0 to default outfile namestr dict (#1031)
2023-04-17 slarenAdd LoRA support (#820)
2023-04-17 Arik Poznanskillama : well-defined static initialization of complex...
2023-04-17 Georgi Gerganovquantize-stats : fix bug in --type argument
2023-04-17 Georgi Gerganovggml : avoid using ggml_fp16_to_fp32() and ggml_fp32_to...
2023-04-17 Ivan KomarovSpeedup the AVX-512 implementation of ggml_vec_dot_q4_0...
2023-04-16 slarenFix: do not close file on mmap (#1017)
2023-04-16 Georgi Gerganovstdout : vertical align outputs for better readibility
2023-04-16 Pavol Rusnakexamples: add missing <ctime> include for time() (...
2023-04-16 nanahiFix msys2 build error and warnings (#1009)
2023-04-15 comexconvert.py: Fix loading safetensors and ggml format...
2023-04-15 Stephan WalterFix potential int8 overflow in non-SIMD vec_dot (#986)
2023-04-15 Stephan WalterRefactor ggml.c for future tensor types (#1001)
2023-04-15 Georgi Gerganovggml : add Q8_0 quantization for intermediate results...
2023-04-15 Georgi Gerganovggml : use posix_memalign on non-Windows env
2023-04-15 Ivan Komarovbenchmark : fix result validation in benchmark-q4_0...
2023-04-15 katsu560cmake : add finding the OpenBLAS header file (#992)
2023-04-14 Pavol RusnakRevert "main : alternative instruct mode (Vicuna suppor...
2023-04-14 Pavol Rusnakpy : bump sentencepiece to 0.1.98 to support Python...
2023-04-14 Stephan Waltermake : fix dependencies, use auto variables (#983)
2023-04-14 Pavol RusnakExpose type name from ggml (#970)
2023-04-14 Tomáš Pazdioramain : alternative instruct mode (Vicuna support, etc...
2023-04-14 Kerfuffleggml : add unary and binary map operations (#874)
2023-04-14 Pavol Rusnakpy : cleanup dependencies (#962)
2023-04-14 Pavol Rusnakpy : fix flake8 and isort nitpicks (#960)
2023-04-14 Georgi Gerganovggml : minor
2023-04-14 Georgi Gerganovggml : always allocate buffers with size multiple of...
2023-04-14 comexpy : new conversion script (#545)
2023-04-14 Georgi Gerganovggml : fix q4_1 dot product types
2023-04-14 Howard Suggml : optimize rope function to avoid call powf in...
2023-04-13 Gary Linscottperplexity : add support for batch size to `--perplexit...
2023-04-13 CRD716common : remove unnecessary includes (#947)
2023-04-13 Georgi Gerganovggml : add GGML_DEFAULT_N_THREADS
2023-04-13 Georgi Gerganovggml : speed-up ggml_vec_dot_q4_1() ARM_NEON + 32-bit...
2023-04-13 Georgi Gerganovllama : merge llama_internal.h into llama.h
2023-04-13 Georgi Gerganovgitignore : benchmark
2023-04-13 Stephan Walterggml : optimize non-SIMD Q4_0 vector dot product (...
2023-04-13 Pavol Rusnakggml : introduce GGML_ALIGNED_MALLOC/GGML_ALIGNED_FREE...
2023-04-13 CRD716fix whitespace (#944)
2023-04-13 CRD716readme : remove python 3.10 warning (#929)
2023-04-13 Genkagaku.GPTreadme : llama node binding (#911)
2023-04-13 Pavol Rusnakflake.nix: add all binaries from bin (#848)
2023-04-13 Juddzig : update build.zig (#872)
2023-04-13 Vladimirggml : update cblas_sgemm columns var to be more reason...
next