Skip to content

Commit

Permalink
Updated to 1.43
Browse files Browse the repository at this point in the history
  • Loading branch information
korewaChino committed Sep 11, 2023
1 parent 25249f3 commit f18afaa
Show file tree
Hide file tree
Showing 3 changed files with 19 additions and 26 deletions.
41 changes: 17 additions & 24 deletions .github/workflows/docker.yml
Original file line number Diff line number Diff line change
Expand Up @@ -3,56 +3,49 @@ name: Build and Push
on:
push:
branches:
- 'main'
- "main"

jobs:
docker-cuda:
runs-on: ubuntu-latest
steps:
-
name: Set up QEMU
- name: Set up QEMU
uses: docker/setup-qemu-action@v2
-
name: Set up Docker Buildx
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
-
name: Login to Docker Hub
- name: Login to Docker Hub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_TOKEN }}
-
name: Build and push
- name: Build and push
uses: docker/build-push-action@v4
with:
push: true
tags: |
korewachino/koboldcpp:latest
korewachino/koboldcpp:1.42.1
korewachino/koboldcpp:cuda
korewachino/koboldcpp:latest
korewachino/koboldcpp:1.43
korewachino/koboldcpp:cuda
docker-nocuda:
runs-on: ubuntu-latest
steps:
-
name: Set up QEMU
- name: Set up QEMU
uses: docker/setup-qemu-action@v2
-
name: Set up Docker Buildx
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
-
name: Login to Docker Hub
- name: Login to Docker Hub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_TOKEN }}
-
name: Build and push
- name: Build and push
uses: docker/build-push-action@v4
with:
file: nocuda.dockerfile
push: true
tags: |
korewachino/koboldcpp:nocuda
korewachino/koboldcpp:1.42.1-rocm
korewachino/koboldcpp:1.42.1-nocuda
korewachino/koboldcpp:rocm
korewachino/koboldcpp:nocuda
korewachino/koboldcpp:1.43-rocm
korewachino/koboldcpp:1.43-nocuda
korewachino/koboldcpp:rocm
2 changes: 1 addition & 1 deletion Dockerfile
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,7 @@ COPY ./koboldcpp ./home/koboldcpp
WORKDIR /home/koboldcpp

RUN pip install --upgrade pip setuptools wheel \
&& pip install -r requirements.txt
&& pip install -r requirements.txt

# You need this environment variable to make sure that the CUDA architecture works for all GPUs
ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
Expand Down
2 changes: 1 addition & 1 deletion koboldcpp
Submodule koboldcpp updated 58 files
+3 −0 .editorconfig
+36 −0 .github/workflows/code-coverage.yml
+5 −0 CMakeLists.txt
+9 −8 Makefile
+11 −2 Package.swift
+14 −0 codecov.yml
+247 −95 common/common.cpp
+37 −1 common/common.h
+1 −0 common/console.cpp
+643 −1 common/log.h
+20 −14 convert-falcon-hf-to-gguf.py
+13 −8 convert-gptneox-hf-to-gguf.py
+0 −258 convert-llama-7b-pth-to-gguf.py
+144 −38 convert-llama-ggml-to-gguf.py
+0 −277 convert-llama-hf-to-gguf.py
+5 −3 convert-lora-to-ggml.py
+108 −80 convert.py
+1 −0 examples/CMakeLists.txt
+0 −5 examples/baby-llama/baby-llama.cpp
+5 −3 examples/beam-search/beam-search.cpp
+38 −35 examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp
+21 −21 examples/gguf/gguf.cpp
+33 −33 examples/gptneox-wip/falcon-main.cpp
+38 −37 examples/gptneox-wip/gptneox-main.cpp
+20 −20 examples/llama-bench/llama-bench.cpp
+2 −2 examples/main/README.md
+26 −129 examples/main/main.cpp
+6 −6 examples/perplexity/perplexity.cpp
+19 −5 examples/quantize/quantize.cpp
+2,005 −1,939 examples/server/index.html.hpp
+37 −4 examples/server/public/index.html
+55 −43 examples/server/server.cpp
+8 −0 examples/speculative/CMakeLists.txt
+292 −0 examples/speculative/speculative.cpp
+4 −1 examples/train-text-from-scratch/convert-train-checkpoint-to-gguf.py
+16 −17 expose.cpp
+99 −21 ggml-alloc.c
+83 −23 ggml-cuda.cu
+49 −13 ggml-metal.m
+160 −74 ggml-metal.metal
+7 −7 ggml-opencl.cpp
+227 −46 ggml.c
+36 −34 gguf-py/gguf/gguf.py
+1 −1 gguf-py/pyproject.toml
+19 −9 gpttype_adapter.cpp
+42 −0 grammars/c.gbnf
+34 −0 grammars/json_arr.gbnf
+45 −12 k_quants.c
+174 −64 klite.embd
+46 −38 koboldcpp.py
+156 −58 llama.cpp
+3 −0 llama.h
+12 −2 model_adapter.cpp
+7 −2 model_adapter.h
+5 −0 mypy.ini
+4 −0 otherarch/llama-util.h
+17 −1 otherarch/llama_v3.cpp
+13 −5 otherarch/rwkv_v3.cpp

0 comments on commit f18afaa

Please sign in to comment.