From 6da711dde1d2bc4ada7685d53b580a39829fd3b3 Mon Sep 17 00:00:00 2001 From: bojiang Date: Thu, 31 Oct 2024 20:29:13 +0800 Subject: [PATCH 1/5] chore: add gen readme CI back --- .github/workflows/gen_readme.yml | 47 ------------------------------- .github/workflows/make_bentos.yml | 38 ++++++++++++++++++++++++- 2 files changed, 37 insertions(+), 48 deletions(-) delete mode 100644 .github/workflows/gen_readme.yml diff --git a/.github/workflows/gen_readme.yml b/.github/workflows/gen_readme.yml deleted file mode 100644 index 04a17cb5..00000000 --- a/.github/workflows/gen_readme.yml +++ /dev/null @@ -1,47 +0,0 @@ -name: Generate README - -on: - push: - branches: - - main - -permissions: - contents: write - -jobs: - generate-readme: - runs-on: ubuntu-latest - - steps: - - name: Checkout repository - uses: actions/checkout@v3 - - - name: Set up Python - uses: actions/setup-python@v4 - with: - python-version: '3.9' - - - name: Install dependencies - run: | - python -m pip install --upgrade pip - pip install jinja2 openllm - - - name: Run gen_readme.py - run: python gen_readme.py - - - name: Configure git - run: | - git config --global user.name 'github-actions[bot]' - git config --global user.email 'github-actions[bot]@users.noreply.github.com' - - - name: Commit and push changes - run: | - if [[ -n $(git status -s) ]]; then - git add README.md - git commit -m "Update README.md" - git push - else - echo "No changes to commit" - fi - env: - GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} diff --git a/.github/workflows/make_bentos.yml b/.github/workflows/make_bentos.yml index ea533874..8bab90cb 100644 --- a/.github/workflows/make_bentos.yml +++ b/.github/workflows/make_bentos.yml @@ -9,7 +9,7 @@ permissions: contents: write jobs: - generate-readme: + make-bentos: runs-on: ubuntu-latest steps: @@ -45,3 +45,39 @@ jobs: fi env: GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} + generate-readme: + runs-on: ubuntu-latest + + steps: + - name: Checkout repository + uses: actions/checkout@v3 + + - name: Set up Python + uses: actions/setup-python@v4 + with: + python-version: '3.9' + + - name: Install dependencies + run: | + python -m pip install --upgrade pip + pip install jinja2 openllm + + - name: Run gen_readme.py + run: python gen_readme.py + + - name: Configure git + run: | + git config --global user.name 'github-actions[bot]' + git config --global user.email 'github-actions[bot]@users.noreply.github.com' + + - name: Commit and push changes + run: | + if [[ -n $(git status -s) ]]; then + git add README.md + git commit -m "Update README.md" + git push + else + echo "No changes to commit" + fi + env: + GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} From 9a0b1f0dbba87082465a14809308891ef8f138a9 Mon Sep 17 00:00:00 2001 From: "github-actions[bot]" Date: Thu, 31 Oct 2024 12:30:10 +0000 Subject: [PATCH 2/5] Update README.md --- README.md | 109 +++++++++++++++++++++++++++--------------------------- 1 file changed, 55 insertions(+), 54 deletions(-) diff --git a/README.md b/README.md index f02268e8..1a19b387 100644 --- a/README.md +++ b/README.md @@ -27,60 +27,61 @@ $ openllm repo update $ openllm model list model version repo required GPU RAM platforms ------------- ------------------------------------------ ------- ------------------ ----------- -codestral codestral:22b-v0.1-fp16-1802 default 80G linux -gemma gemma:2b-instruct-fp16-f738 default 12G linux - gemma:7b-instruct-fp16-7ca0 default 24G linux - gemma:7b-instruct-awq-4bit-b214 default 12G linux -gemma2 gemma2:9b-instruct-fp16-e0c2 default 24G linux - gemma2:27b-instruct-fp16-8ee8 default 80G linux -jamba1.5 jamba1.5:mini-fp16-f103 default 80Gx4 linux -llama2 llama2:7b-chat-fp16-d7da default 16G linux - llama2:7b-chat-awq-4bit-bad5 default 12G linux - llama2:13b-chat-fp16-01e5 default 40G linux - llama2:70b-chat-fp16-fb96 default 80Gx2 linux -llama3 llama3:8b-instruct-fp16-c11d default 24G linux - llama3:8b-instruct-awq-4bit-4c9c default 12G linux - llama3:70b-instruct-fp16-571d default 80Gx2 linux - llama3:70b-instruct-awq-4bit-a15f default 80G linux -llama3.1 llama3.1:8b-instruct-fp16-79d3 default 24G linux - llama3.1:8b-instruct-awq-4bit-522a default 12G linux - llama3.1:70b-instruct-fp16-859e default 80Gx2 linux - llama3.1:70b-instruct-awq-4bit-7cac default 80G linux - llama3.1:405b-instruct-awq-4bit-d543 default 80Gx4 linux -llama3.2 llama3.2:1b-instruct-fp16-dc66 default 12G linux - llama3.2:1b-instruct-ggml-fp16-linux-3694 default linux - llama3.2:1b-instruct-ggml-fp16-darwin-9fcb default macos - llama3.2:3b-instruct-fp16-6cc3 default 12G linux - llama3.2:11b-vision-instruct-1459 default 80G linux -mistral mistral:7b-instruct-fp16-31e6 default 24G linux - mistral:7b-instruct-awq-4bit-0506 default 12G linux - mistral:24b-instruct-nemo-9505 default 80G linux -mistral-large mistral-large:123b-instruct-fp16-dd43 default 80Gx4 linux - mistral-large:123b-instruct-awq-4bit-c5c3 default 80G linux -mixtral mixtral:8x7b-instruct-v0.1-fp16-9bac default 80Gx2 linux - mixtral:8x7b-instruct-v0.1-awq-4bit-3eaf default 40G linux -phi3 phi3:3.8b-instruct-fp16-7cdd default 12G linux - phi3:3.8b-instruct-ggml-q4-9c53 default macos -pixtral pixtral:12b-240910-3480 default 80G linux -qwen2 qwen2:0.5b-instruct-fp16-603d default 12G linux - qwen2:1.5b-instruct-fp16-8219 default 12G linux - qwen2:7b-instruct-fp16-15ec default 24G linux - qwen2:7b-instruct-awq-4bit-bb90 default 12G linux - qwen2:57b-a14b-instruct-fp16-da51 default 80Gx2 linux - qwen2:72b-instruct-fp16-02ba default 80Gx2 linux - qwen2:72b-instruct-awq-4bit-40e4 default 80G linux -qwen2.5 qwen2.5:0.5b-instruct-fp16-5ae4 default 12G linux - qwen2.5:1.5b-instruct-fp16-0007 default 12G linux - qwen2.5:3b-instruct-fp16-46fb default 12G linux - qwen2.5:7b-instruct-fp16-bc18 default 24G linux - qwen2.5:14b-instruct-fp16-6fe3 default 80G linux - qwen2.5:14b-instruct-ggml-q4-darwin-5997 default macos - qwen2.5:14b-instruct-ggml-q8-darwin-d0e8 default macos - qwen2.5:32b-instruct-fp16-7848 default 80G linux - qwen2.5:32b-instruct-awq-4bit-2438 default 40G linux - qwen2.5:32b-instruct-ggml-fp16-darwin-00e1 default macos - qwen2.5:72b-instruct-fp16-864f default 80Gx2 linux - qwen2.5:72b-instruct-ggml-q4-darwin-8b90 default macos +codestral codestral:22b-v0.1-fp16-2638 default 80G linux +gemma gemma:2b-instruct-fp16-d4c6 default 12G linux + gemma:7b-instruct-fp16-b15c default 24G linux + gemma:7b-instruct-awq-4bit-464a default 12G linux +gemma2 gemma2:9b-instruct-fp16-b9d3 default 24G linux + gemma2:27b-instruct-fp16-9f48 default 80G linux +jamba1.5 jamba1.5:mini-fp16-9c32 default 80Gx4 linux +llama2 llama2:7b-chat-fp16-3523 default 16G linux + llama2:7b-chat-awq-4bit-a536 default 12G linux + llama2:13b-chat-fp16-b2a0 default 40G linux + llama2:70b-chat-fp16-7fe8 default 80Gx2 linux +llama3 llama3:8b-instruct-fp16-07fd default 24G linux + llama3:8b-instruct-awq-4bit-da0a default 12G linux + llama3:70b-instruct-fp16-4863 default 80Gx2 linux + llama3:70b-instruct-awq-4bit-c17d default 80G linux +llama3.1 llama3.1:8b-instruct-fp16-d75d default 24G linux + llama3.1:8b-instruct-awq-4bit-96eb default 12G linux + llama3.1:70b-instruct-fp16-b86d default 80Gx2 linux + llama3.1:70b-instruct-awq-4bit-dbcc default 80G linux + llama3.1:405b-instruct-awq-4bit-2358 default 80Gx4 linux +llama3.2 llama3.2:1b-instruct-fp16-62c6 default 12G linux + llama3.2:1b-instruct-ggml-fp16-linux-60fa default linux + llama3.2:1b-instruct-ggml-fp16-darwin-8d35 default macos + llama3.2:3b-instruct-fp16-53eb default 12G linux + llama3.2:11b-vision-instruct-8926 default 80G linux +mistral mistral:7b-instruct-fp16-6aea default 24G linux + mistral:7b-instruct-awq-4bit-1bf8 default 12G linux + mistral:24b-instruct-nemo-9adb default 80G linux +mistral-large mistral-large:123b-instruct-fp16-9008 default 80Gx4 linux + mistral-large:123b-instruct-awq-4bit-377e default 80G linux +mixtral mixtral:8x7b-instruct-v0.1-fp16-b8de default 80Gx2 linux + mixtral:8x7b-instruct-v0.1-awq-4bit-1392 default 40G linux +phi3 phi3:3.8b-instruct-fp16-ac3c default 12G linux + phi3:3.8b-instruct-ggml-q4-463e default macos +pixtral pixtral:12b-240910-82ad default 80G linux +qwen2 qwen2:0.5b-instruct-fp16-fe8a default 12G linux + qwen2:1.5b-instruct-fp16-74cd default 12G linux + qwen2:7b-instruct-fp16-e8fa default 24G linux + qwen2:7b-instruct-awq-4bit-dc8b default 12G linux + qwen2:57b-a14b-instruct-fp16-55df default 80Gx2 linux + qwen2:72b-instruct-fp16-5ba6 default 80Gx2 linux + qwen2:72b-instruct-awq-4bit-2907 default 80G linux +qwen2.5 qwen2.5:0.5b-instruct-fp16-d59e default 12G linux + qwen2.5:1.5b-instruct-fp16-d1e9 default 12G linux + qwen2.5:3b-instruct-fp16-b569 default 12G linux + qwen2.5:7b-instruct-fp16-60be default 24G linux + qwen2.5:14b-instruct-fp16-9539 default 80G linux + qwen2.5:14b-instruct-ggml-q4-darwin-1cf2 default macos + qwen2.5:14b-instruct-ggml-q8-darwin-f06a default macos + qwen2.5:32b-instruct-fp16-7b67 default 80G linux + qwen2.5:32b-instruct-awq-4bit-6958 default 40G linux + qwen2.5:32b-instruct-ggml-fp16-darwin-809c default macos + qwen2.5:72b-instruct-fp16-aa9c default 80Gx2 linux + qwen2.5:72b-instruct-ggml-q4-darwin-a138 default macos +qwen2vl qwen2vl:7b-instruct-fp16-8a12 default 24G linux ``` From ffd75a876b7eb6406560f1942ee1decd89500ee1 Mon Sep 17 00:00:00 2001 From: bojiang Date: Thu, 31 Oct 2024 20:30:42 +0800 Subject: [PATCH 3/5] fix: gen readme after build bento --- .github/workflows/make_bentos.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/make_bentos.yml b/.github/workflows/make_bentos.yml index 8bab90cb..fc3a22e8 100644 --- a/.github/workflows/make_bentos.yml +++ b/.github/workflows/make_bentos.yml @@ -46,6 +46,7 @@ jobs: env: GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} generate-readme: + needs: make-bentos runs-on: ubuntu-latest steps: From 3f97508b6e3cd7cadc1d4625cf953457a01889d8 Mon Sep 17 00:00:00 2001 From: bojiang Date: Thu, 31 Oct 2024 20:36:12 +0800 Subject: [PATCH 4/5] fix: unify gen readme and make bentos --- .github/workflows/make_bentos.yml | 35 +++---------------------------- gen_readme.py | 2 -- 2 files changed, 3 insertions(+), 34 deletions(-) diff --git a/.github/workflows/make_bentos.yml b/.github/workflows/make_bentos.yml index fc3a22e8..85b4785a 100644 --- a/.github/workflows/make_bentos.yml +++ b/.github/workflows/make_bentos.yml @@ -29,35 +29,6 @@ jobs: make cd .. - - name: Configure git - run: | - git config --global user.name 'github-actions[bot]' - git config --global user.email 'github-actions[bot]@users.noreply.github.com' - - - name: Commit and push changes - run: | - if [[ -n $(git status -s) ]]; then - git add bentoml/bentos/* - git commit -m "chore: make bentos" - git push origin nightly - else - echo "No changes to commit" - fi - env: - GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }} - generate-readme: - needs: make-bentos - runs-on: ubuntu-latest - - steps: - - name: Checkout repository - uses: actions/checkout@v3 - - - name: Set up Python - uses: actions/setup-python@v4 - with: - python-version: '3.9' - - name: Install dependencies run: | python -m pip install --upgrade pip @@ -74,9 +45,9 @@ jobs: - name: Commit and push changes run: | if [[ -n $(git status -s) ]]; then - git add README.md - git commit -m "Update README.md" - git push + git add bentoml/bentos/* + git commit -m "chore: make bentos" + git push origin nightly else echo "No changes to commit" fi diff --git a/gen_readme.py b/gen_readme.py index d6adf3c6..83c0a6ea 100644 --- a/gen_readme.py +++ b/gen_readme.py @@ -3,8 +3,6 @@ import sys -subprocess.check_call([sys.executable, "-m", "openllm", "repo", "remove", "default"]) -subprocess.check_call([sys.executable, "-m", "openllm", "repo", "update"]) model_list = subprocess.check_output( [sys.executable, "-m", "openllm", "model", "list"], env={"OPENLLM_TEST_REPO": "."}, From 27de4446ebd6bb393cd2f464fd14adcb88039f95 Mon Sep 17 00:00:00 2001 From: bojiang Date: Thu, 31 Oct 2024 20:38:13 +0800 Subject: [PATCH 5/5] fix: gen readme --- .github/workflows/make_bentos.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/.github/workflows/make_bentos.yml b/.github/workflows/make_bentos.yml index 85b4785a..390878e7 100644 --- a/.github/workflows/make_bentos.yml +++ b/.github/workflows/make_bentos.yml @@ -46,6 +46,7 @@ jobs: run: | if [[ -n $(git status -s) ]]; then git add bentoml/bentos/* + git add README.md git commit -m "chore: make bentos" git push origin nightly else