cosmopolitan/third_party/ggml
2023-06-03 10:29:12 -07:00
..
common.cc Make more ML improvements 2023-05-16 08:07:23 -07:00
common.h Make more ML improvements 2023-05-16 08:07:23 -07:00
companionai.txt Upgrade llama.cpp to e6a46b0ed1884c77267dc70693183e3b7164e0e0 2023-05-10 04:20:48 -07:00
fp16.c Make more ML improvements 2023-05-16 08:07:23 -07:00
fp16.h Make more ML improvements 2023-05-16 08:07:23 -07:00
fp16.internal.h Make more ML improvements 2023-05-16 08:07:23 -07:00
ggjt.v1.c Get radpajama to build 2023-05-13 20:44:36 -07:00
ggjt.v1.internal.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q4_0.c Make more ML improvements 2023-05-16 08:07:23 -07:00
ggjt.v1.q4_0.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q4_1.c Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q4_1.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q4_2.c Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q4_2.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q5_0.c Fix the AARCH64 build 2023-05-13 08:19:44 -07:00
ggjt.v1.q5_0.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q5_1.c Fix the AARCH64 build 2023-05-13 08:19:44 -07:00
ggjt.v1.q5_1.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q8_0.c Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q8_0.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q8_1.c Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggjt.v1.q8_1.h Add support for new GGJT v2 quantizers 2023-05-13 08:08:32 -07:00
ggml.c Sync llama.cpp to 6986c7835adc13ba3f9d933b95671bb1f3984dc6 2023-06-03 10:29:12 -07:00
ggml.h Sync llama.cpp to 6986c7835adc13ba3f9d933b95671bb1f3984dc6 2023-06-03 10:29:12 -07:00
ggml.mk Make more ML improvements 2023-05-16 08:07:23 -07:00
LICENSE Import llama.cpp 2023-04-27 14:37:14 -07:00
llama.cc Sync llama.cpp to 6986c7835adc13ba3f9d933b95671bb1f3984dc6 2023-06-03 10:29:12 -07:00
llama.h Make more ML improvements 2023-05-16 08:07:23 -07:00
llama_util.h Make more ML improvements 2023-05-16 08:07:23 -07:00
main.cc Make more ML improvements 2023-05-16 08:07:23 -07:00
perplexity.cc Perform some code cleanup 2023-05-15 16:32:10 -07:00
quantize.cc Introduce native support for MacOS ARM64 2023-05-20 04:17:03 -07:00
README.cosmo Make more ML improvements 2023-05-16 08:07:23 -07:00

DESCRIPTION

  ggml is a machine learning library useful for LLM inference on CPUs

LICENSE

  MIT

ORIGIN

  https://github.com/ggerganov/llama.cpp
  commit 0b2da20538d01926b77ea237dd1c930c4d20b686
  Author: Stephan Walter <stephan@walter.name>
  Date:   Wed Apr 26 20:26:42 2023 +0000
  ggml : slightly faster AVX2 implementation for Q5 (#1197)

LOCAL CHANGES

  - Make it possible for loaded prompts to be cached to disk
  - Introduce -v and --verbose flags
  - Reduce batch size from 512 to 32
  - Allow --n_keep to specify a substring of prompt
  - Don't print stats / diagnostics unless -v is passed
  - Reduce --top_p default from 0.95 to 0.70
  - Change --reverse-prompt to no longer imply --interactive
  - Permit --reverse-prompt specifying custom EOS if non-interactive
  - Refactor headers per cosmo convention
  - Replace code like 'ggjt' with READ32BE("ggjt")
  - Remove C++ exceptions; use Die() function instead
  - Removed division from matrix multiplication.
  - Let quantizer convert between ggmt formats