Merge branch 'master' into cb-by-default

.github/workflows/causal_lm_cpp.yml

@@ -53,17 +53,17 @@ jobs:
           wget https://huggingface.co/smangrul/tinyllama_lora_sql/resolve/main/adapter_model.safetensors?download=true -O adapter_model.safetensors
       - run: >
           . ./ov/setupvars.sh
-          && timeout 25s ./build/samples/cpp/multinomial_causal_lm/multinomial_causal_lm ./open_llama_3b_v2/ a
+          && timeout 35s ./build/samples/cpp/multinomial_causal_lm/multinomial_causal_lm ./open_llama_3b_v2/ a
         env:
           PYTHONPATH: "./build"
       - run: >
           . ./ov/setupvars.sh
-          && timeout 25s ./samples/python/multinomial_causal_lm/multinomial_causal_lm.py ./open_llama_3b_v2/ b
+          && timeout 35s ./samples/python/multinomial_causal_lm/multinomial_causal_lm.py ./open_llama_3b_v2/ b
         env:
           PYTHONPATH: "./build"
       - run: >
           . ./ov/setupvars.sh
-          && timeout 25s ./build/samples/cpp/text_generation/greedy_causal_lm ./open_llama_3b_v2/ "return 0"
+          && timeout 35s ./build/samples/cpp/text_generation/greedy_causal_lm ./open_llama_3b_v2/ "return 0"
           | diff <(timeout 25s samples/python/text_generation/greedy_causal_lm.py ./open_llama_3b_v2/ "return 0") -
         env:
           PYTHONPATH: "./build"

.github/workflows/genai-tools.yml

@@ -44,7 +44,7 @@ jobs:
       with:
         platform: ubuntu22
         commit_packages_to_provide: wheels
-        revision: 345163f87953fb0dd8dd590257eb7fc84378da8e
+        revision: latest_available_commit
 
   llm_bench:
     name: 'LLM bench tests'

.github/workflows/linux.yml

@@ -52,7 +52,7 @@ jobs:
       with:
         platform: ubuntu22
         commit_packages_to_provide: wheels
-        revision: 345163f87953fb0dd8dd590257eb7fc84378da8e
+        revision: latest_available_commit
 
     - name: Clone docker tag from OpenVINO repo
       uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2

.github/workflows/mac.yml

@@ -17,7 +17,7 @@ concurrency:
 
 env:
   PYTHON_VERSION: '3.10'
-  OV_BRANCH: 345163f87953fb0dd8dd590257eb7fc84378da8e
+  OV_BRANCH: 'master'
   OV_TARBALL: ''
 
 jobs:

.github/workflows/stable_diffusion_1_5_cpp.yml

@@ -45,7 +45,7 @@ jobs:
       with:
         platform: ubuntu22
         commit_packages_to_provide: wheels
-        revision: 345163f87953fb0dd8dd590257eb7fc84378da8e
+        revision: latest_available_commit
 
   openvino_download_windows:
     name: Download OpenVINO for Windows
@@ -71,7 +71,7 @@ jobs:
       with:
         platform: windows
         commit_packages_to_provide: wheels
-        revision: 345163f87953fb0dd8dd590257eb7fc84378da8e
+        revision: latest_available_commit
 
   stable_diffusion_1_5_cpp-linux:
     runs-on: ubuntu-22.04-8-cores

.github/workflows/windows.yml

@@ -17,7 +17,7 @@ concurrency:
 
 env:
   PYTHON_VERSION: '3.11'
-  OV_BRANCH: 345163f87953fb0dd8dd590257eb7fc84378da8e
+  OV_BRANCH: 'master'
   OV_TARBALL: ''
 
 jobs:
@@ -310,6 +310,12 @@ jobs:
           . "${{ env.OV_INSTALL_DIR }}/setupvars.ps1"
           python -m pip install . --verbose --find-links ${env:OV_INSTALL_DIR}/wheels
           python -m pip install ./tools/who_what_benchmark --find-links ${env:OV_INSTALL_DIR}/wheels
+          
+          # will install transformers 4.46.3 version
+          # transformers 4.46.3 will enable return_timestamps tests
+          # this check enabled for windows only. Ticket: 160205.
+          python -m pip install git+https://github.com/huggingface/optimum-intel.git@753f84db6e0966580eb9eaa74a808213be730631
+          
           python -m pytest -v ./tests/python_tests/test_whisper_pipeline.py -k "not test_smoke"
 
   genai_python_lib_vlm:

README.md

@@ -73,9 +73,9 @@ optimum-cli export openvino --model "TinyLlama/TinyLlama-1.1B-Chat-v1.0" --weigh
 ### Run generation using LLMPipeline API in Python
 
 ```python
-import openvino_genai as ov_genai
+import openvino_genai
 #Will run model on CPU, GPU or NPU are possible options
-pipe = ov_genai.LLMPipeline("./TinyLlama-1.1B-Chat-v1.0/", "CPU")
+pipe = openvino_genai.LLMPipeline("./TinyLlama-1.1B-Chat-v1.0/", "CPU")
 print(pipe.generate("The Sun is yellow because", max_new_tokens=100))
 ```
 
@@ -128,11 +128,11 @@ curl -O "https://storage.openvinotoolkit.org/test_data/images/dog.jpg"
 ```python
 import numpy as np
 import openvino as ov
-import openvino_genai as ov_genai
+import openvino_genai
 from PIL import Image
 
 # Choose GPU instead of CPU in the line below to run the model on Intel integrated or discrete GPU
-pipe = ov_genai.VLMPipeline("./InternVL2-1B", "CPU")
+pipe = openvino_genai.VLMPipeline("./InternVL2-1B", "CPU")
 pipe.start_chat()
 
 image = Image.open("dog.jpg")

SUPPORTED_MODELS.md

@@ -14,7 +14,6 @@
       <td>ChatGLM</td>
       <td>
         <ul>
-          <li><a href="https://huggingface.co/THUDM/chatglm2-6b"><code>THUDM/chatglm2-6b</code></a></li>
           <li><a href="https://huggingface.co/THUDM/chatglm3-6b"><code>THUDM/chatglm3-6b</code></a></li>
         </ul>
       </td>

samples/export-requirements.txt

@@ -2,7 +2,7 @@
 --extra-index-url https://storage.openvinotoolkit.org/simple/wheels/pre-release
 --extra-index-url https://storage.openvinotoolkit.org/simple/wheels/nightly
 openvino-tokenizers~=2025.0.0.0.dev
-optimum-intel @ git+https://github.com/huggingface/optimum-intel.git@753f84db6e0966580eb9eaa74a808213be730631
+optimum-intel @ git+https://github.com/huggingface/optimum-intel.git
 numpy<2.0.0; sys_platform == 'darwin'
 einops==0.8.0  # For Qwen
 transformers_stream_generator==0.0.5  # For Qwen

tests/python_tests/models/real_models

@@ -27,7 +27,6 @@ Salesforce/codegen-350M-multi
 Salesforce/codegen-350M-nl
 Salesforce/codegen2-1b
 # Salesforce/xgen-7b-8k-base: Transformers issue - Object of type method is not JSON serializable (https://huggingface.co/Salesforce/xgen-7b-8k-base/discussions/32)
-THUDM/chatglm2-6b
 THUDM/chatglm3-6b
 TheBloke/Wizard-Vicuna-30B-Uncensored-GPTQ
 TinyLlama/TinyLlama-1.1B-Chat-v0.6

tests/python_tests/ov_genai_test_utils.py

@@ -26,7 +26,7 @@ def get_models_list():
         "facebook/opt-125m",
         "microsoft/phi-1_5",
         "microsoft/phi-2",
-        "THUDM/chatglm2-6b",
+        "THUDM/chatglm3-6b",
         "Qwen/Qwen2-0.5B-Instruct",
         "Qwen/Qwen-7B-Chat",
         "Qwen/Qwen1.5-7B-Chat",

tests/python_tests/requirements.txt

@@ -1,6 +1,6 @@
 --extra-index-url https://download.pytorch.org/whl/cpu
 diffusers==0.32.1
-optimum-intel @ git+https://github.com/huggingface/optimum-intel.git@753f84db6e0966580eb9eaa74a808213be730631
+optimum-intel @ git+https://github.com/huggingface/optimum-intel.git
 numpy<2.0.0; platform_system == "Darwin" and platform_machine == "x86_64"
 onnx==1.17.0
 pytest