merge fix

.github/workflows/action_gpu_basic_tests.yml

@@ -43,7 +43,7 @@ jobs:
       - name: Ensure NVIDIA SDK available
         run: |
           sudo apt-get -y install cuda-toolkit
-          echo "/usr/local/cuda-12.5/bin" >> $GITHUB_PATH
+          echo "/usr/local/cuda-12.6/bin" >> $GITHUB_PATH
       - name: Install dependencies
         shell: bash
         run: |

.github/workflows/action_plain_basic_tests.yml

@@ -48,8 +48,8 @@ jobs:
         shell: bash
         run: |
           pytest --cov=guidance --cov-report=xml --cov-report=term-missing \
-          --selected_model ${{ inputs.model }} \
-          ./tests/unit ./tests/model_integration ./tests/model_specific
+            --selected_model ${{ inputs.model }} \
+            ./tests/unit ./tests/model_integration ./tests/model_specific
       - name: Upload coverage reports to Codecov
         uses: codecov/codecov-action@v4
         with:

.github/workflows/ci_tests.yml

@@ -47,7 +47,7 @@ jobs:
       - name: Ensure NVIDIA SDK available
         run: |
           sudo apt-get -y install cuda-toolkit
-          echo "/usr/local/cuda-12.5/bin" >> $GITHUB_PATH
+          echo "/usr/local/cuda-12.6/bin" >> $GITHUB_PATH
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
@@ -75,10 +75,10 @@ jobs:
         env:
           HF_TOKEN: ${{ secrets.HF_TOKEN }}
           # Configure endpoints for Azure OpenAI
-          AZUREAI_CHAT_ENDPOINT: ${{ secrets.AZUREAI_CHAT_ENDPOINT }}
-          AZUREAI_CHAT_MODEL: ${{ secrets.AZUREAI_CHAT_MODEL }}
-          AZUREAI_COMPLETION_ENDPOINT: ${{ secrets.AZUREAI_COMPLETION_ENDPOINT }}
-          AZUREAI_COMPLETION_MODEL: ${{ secrets.AZUREAI_COMPLETION_MODEL }}
+          AZUREAI_CHAT_ENDPOINT: ${{ vars.AZUREAI_CHAT_ENDPOINT }}
+          AZUREAI_CHAT_MODEL: ${{ vars.AZUREAI_CHAT_MODEL }}
+          AZUREAI_COMPLETION_ENDPOINT: ${{ vars.AZUREAI_COMPLETION_ENDPOINT }}
+          AZUREAI_COMPLETION_MODEL: ${{ vars.AZUREAI_COMPLETION_MODEL }}
           # Configure endpoints for Azure AI Studio
           AZURE_AI_STUDIO_PHI3_ENDPOINT: ${{ vars.AZURE_AI_STUDIO_PHI3_ENDPOINT }}
           AZURE_AI_STUDIO_PHI3_DEPLOYMENT: ${{ vars.AZURE_AI_STUDIO_PHI3_DEPLOYMENT }}
@@ -132,7 +132,7 @@ jobs:
       - name: Ensure NVIDIA SDK available
         run: |
           sudo apt-get -y install cuda-toolkit
-          echo "/usr/local/cuda-12.5/bin" >> $GITHUB_PATH
+          echo "/usr/local/cuda-12.6/bin" >> $GITHUB_PATH
       - name: Install dependencies
         shell: bash
         run: |
@@ -150,6 +150,9 @@ jobs:
         run: |
           pip install accelerate
           pip uninstall -y llama-cpp-python
+          echo "======================"
+          nvcc --version
+          echo "======================"
           CMAKE_ARGS="-DGGML_CUDA=on" pip install "llama-cpp-python!=0.2.58,!=0.2.75"
       - name: Check GPU available
         run: |

.github/workflows/notebook_tests.yml

@@ -50,7 +50,7 @@ jobs:
       - name: Ensure NVIDIA SDK available
         run: |
           sudo apt-get -y install cuda-toolkit
-          echo "/usr/local/cuda-12.5/bin" >> $GITHUB_PATH
+          echo "/usr/local/cuda-12.6/bin" >> $GITHUB_PATH
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
@@ -78,10 +78,10 @@ jobs:
         env:
           HF_TOKEN: ${{ secrets.HF_TOKEN }}
           # Configure endpoints for Azure OpenAI
-          AZUREAI_CHAT_ENDPOINT: ${{ secrets.AZUREAI_CHAT_ENDPOINT }}
-          AZUREAI_CHAT_MODEL: ${{ secrets.AZUREAI_CHAT_MODEL }}
-          AZUREAI_COMPLETION_ENDPOINT: ${{ secrets.AZUREAI_COMPLETION_ENDPOINT }}
-          AZUREAI_COMPLETION_MODEL: ${{ secrets.AZUREAI_COMPLETION_MODEL }}
+          AZUREAI_CHAT_ENDPOINT: ${{ vars.AZUREAI_CHAT_ENDPOINT }}
+          AZUREAI_CHAT_MODEL: ${{ vars.AZUREAI_CHAT_MODEL }}
+          AZUREAI_COMPLETION_ENDPOINT: ${{ vars.AZUREAI_COMPLETION_ENDPOINT }}
+          AZUREAI_COMPLETION_MODEL: ${{ vars.AZUREAI_COMPLETION_MODEL }}
           # Configure endpoints for Azure AI Studio
           AZURE_AI_STUDIO_PHI3_ENDPOINT: ${{ vars.AZURE_AI_STUDIO_PHI3_ENDPOINT }}
           AZURE_AI_STUDIO_PHI3_DEPLOYMENT: ${{ vars.AZURE_AI_STUDIO_PHI3_DEPLOYMENT }}

.github/workflows/pypi_upload.yml

@@ -16,6 +16,10 @@ jobs:
     runs-on: ubuntu-latest
     outputs:
       include: ${{ steps.set-matrix.outputs.include }}
+
+    env:
+      CIBW_SKIP: pp*
+
     steps:
       - uses: actions/checkout@v4
       - name: Install cibuildwheel
@@ -50,6 +54,7 @@ jobs:
       CIBW_BUILD_VERBOSITY: 1
       CIBW_BEFORE_ALL_LINUX: "curl https://sh.rustup.rs -sSf | sh -s -- --default-toolchain stable -y"
       CIBW_ENVIRONMENT: 'PATH="$PATH:$HOME/.cargo/bin"'
+      CIBW_SKIP: pp*
 
     steps:
       - name: Check out the repo
@@ -154,8 +159,8 @@ jobs:
         name: Install guidance from wheels
       - run: pip install transformers torch
         name: Other installs
-      - run: python -c "import guidance; lm = guidance.models.Transformers('gpt2'); lm += '1,2,3,4,5,' + guidance.gen('num', max_tokens=5, temperature=0); print(f'\n{str(lm)=}\n'); assert lm['num'].startswith('6')"
-        name: Run smoke test
+      # - run: python -c "import guidance; import transformers; lm = guidance.models.Transformers('gpt2'); lm += '1,2,3,4,5,' + guidance.gen('num', max_tokens=5, temperature=0); print(f'\n Transformers Version:{transformers.__version__}\n\n{str(lm)=}\n'); assert lm['num'].startswith('6')"
+      #   name: Run smoke test
 
   publish_wheels:
     permissions:

.github/workflows/workflow-pr-gate.yml

@@ -169,7 +169,7 @@ jobs:
       matrix:
         python-version: ["3.9", "3.10", "3.11", "3.12"]
         model:
-          - "transformers_gpt2_cpu"
+          # - "transformers_gpt2_cpu" See #965
           # - "transformers_phi2_cpu" Seems to get stuck
           # - "transformers_mistral_7b_cpu" See Issue 713
           # - "llamacpp_llama2_7b_cpu" Getting stuck with llama-cpp-python 0.2.77

guidance/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.1.15"
+__version__ = "0.2.0rc1"
 
 import sys
 import types