Skip to content

Commit

Permalink
[CI] ggml: add llama2 7b chat examples
Browse files Browse the repository at this point in the history
Signed-off-by: hydai <[email protected]>
  • Loading branch information
hydai committed Feb 1, 2024
1 parent fc9a144 commit 80be133
Showing 1 changed file with 44 additions and 3 deletions.
47 changes: 44 additions & 3 deletions .github/workflows/llama.yml
Original file line number Diff line number Diff line change
Expand Up @@ -42,7 +42,7 @@ jobs:
VERSION=0.13.5
curl -sSf https://raw.githubusercontent.com/WasmEdge/WasmEdge/master/utils/install.sh | sudo bash -s -- -v $VERSION --plugins wasi_nn-ggml -p /usr/local
- name: Example
- name: Tiny Llama
run: |
cd wasmedge-ggml-llama-interactive
curl -LO https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v0.3-GGUF/resolve/main/tinyllama-1.1b-chat-v0.3.Q5_K_M.gguf
Expand All @@ -56,8 +56,33 @@ jobs:
target/wasm32-wasi/release/wasmedge-ggml-llama-interactive.wasm \
default \
'<|im_start|>system\nYou are an AI assistant<|im_end|>\n<|im_start|>user\nWhere is the capital of Japan?<|im_end|>\n<|im_start|>assistant'
- name: llama2 7b
run: |
cd wasmedge-ggml-llama-interactive
curl -LO https://huggingface.co/TheBloke/Llama-2-7b-Chat-GGUF/resolve/main/llama-2-7b-chat.Q5_K_M.gguf
cargo build --target wasm32-wasi --release
wasmedge --dir .:. \
--nn-preload default:GGML:AUTO:llama-2-7b-chat.Q5_K_M.gguf \
--env enable_log=true \
--env ctx_size=4096 \
--env stream_stdout=true \
--env n_gpu_layers=0 \
target/wasm32-wasi/release/wasmedge-ggml-llama-interactive.wasm \
default \
'[INST] <<SYS>>\nYou are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you do not know the answer to a question, please do not share false information.\n<</SYS>>\nWhat is the capital of Japan?[/INST]'
macos:
runs-on: macos-14
strategy:
matrix:
include:
- name: MacOS-13
host_runner: macos-13
- name: MacOS-14
host_runner: macos-14
name: ${{ matrix.name }}
runs-on: ${{ matrix.host_runner }}
steps:
- uses: actions/checkout@v4

Expand All @@ -71,7 +96,7 @@ jobs:
VERSION=0.13.5
curl -sSf https://raw.githubusercontent.com/WasmEdge/WasmEdge/master/utils/install.sh | sudo bash -s -- -v $VERSION --plugins wasi_nn-ggml -p /usr/local
- name: Example
- name: Tiny Llama
run: |
cd wasmedge-ggml-llama-interactive
curl -LO https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v0.3-GGUF/resolve/main/tinyllama-1.1b-chat-v0.3.Q5_K_M.gguf
Expand All @@ -85,3 +110,19 @@ jobs:
target/wasm32-wasi/release/wasmedge-ggml-llama-interactive.wasm \
default \
'<|im_start|>system\nYou are an AI assistant<|im_end|>\n<|im_start|>user\nWhere is the capital of Japan?<|im_end|>\n<|im_start|>assistant'
- name: llama2 7b
run: |
cd wasmedge-ggml-llama-interactive
curl -LO https://huggingface.co/TheBloke/Llama-2-7b-Chat-GGUF/resolve/main/llama-2-7b-chat.Q5_K_M.gguf
cargo build --target wasm32-wasi --release
wasmedge --dir .:. \
--nn-preload default:GGML:AUTO:llama-2-7b-chat.Q5_K_M.gguf \
--env enable_log=true \
--env ctx_size=4096 \
--env stream_stdout=true \
--env n_gpu_layers=0 \
target/wasm32-wasi/release/wasmedge-ggml-llama-interactive.wasm \
default \
'[INST] <<SYS>>\nYou are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you do not know the answer to a question, please do not share false information.\n<</SYS>>\nWhat is the capital of Japan?[/INST]'

0 comments on commit 80be133

Please sign in to comment.