GHA: update workflows (#1720)

- Align archives with ones, updated in PR
7998955
- Use macos-13 instead of macos-12 as last one is not available
- Ubuntu 20.04 is also deprecated (see migration commit in OV
openvinotoolkit/openvino@9e30651)

Author: ilya-lavrenov

.github/workflows/bandit.yml

@@ -7,7 +7,7 @@ on:
 permissions: read-all  # Required by https://github.com/ossf/scorecard/blob/e23b8ad91fd6a64a0a971ca4fc0a4d1650725615/docs/checks.md#token-permissions
 jobs:
   bandit:
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     steps:
       - uses: actions/checkout@v4
       - uses: actions/setup-python@v4

.github/workflows/causal_lm_cpp.yml

@@ -23,7 +23,7 @@ env:
 
 jobs:
   cpp-multinomial-greedy_causal_lm-ubuntu:
-    runs-on: ubuntu-20.04-8-cores
+    runs-on: ubuntu-22.04-8-cores
     defaults:
       run:
         shell: bash
@@ -82,7 +82,7 @@ jobs:
             ./build/samples/cpp/text_generation/beam_search_causal_lm,
             python ./samples/python/text_generation/beam_search_causal_lm.py,
           ]
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     defaults:
       run:
         shell: bash
@@ -298,7 +298,7 @@ jobs:
           && python samples\python\text_generation\lora.py .\TinyLlama\TinyLlama-1.1B-intermediate-step-1431k-3T\ adapter_model.safetensors "How to create a table with two columns, one of them has type float, another one has type int?"
 
   cpp-greedy_causal_lm-Qwen-7B-Chat:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -332,7 +332,7 @@ jobs:
           PYTHONPATH: "./build"
 
   cpp-beam_search_causal_lm-Qwen1_5-7B-Chat:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -367,7 +367,7 @@ jobs:
           PYTHONPATH: "./build"
 
   cpp-beam_search_causal_lm-Phi-2:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -402,7 +402,7 @@ jobs:
           PYTHONPATH: "./build"
 
   cpp-beam_search_causal_lm-notus-7b-v1:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -437,7 +437,7 @@ jobs:
           PYTHONPATH: "./build"
 
   cpp-speculative_decoding_lm-ubuntu:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -451,7 +451,7 @@ jobs:
       - name: Install OpenVINO
         run: |
           mkdir ./ov/
-          curl ${{ env.l_ov_link }} | tar --directory ./ov/ --strip-components 1 -xz
+          curl ${{ env.l_u22_ov_link }} | tar --directory ./ov/ --strip-components 1 -xz
           sudo ./ov/install_dependencies/install_openvino_dependencies.sh
       - name: Build app
         run: |
@@ -494,7 +494,7 @@ jobs:
           LD_LIBRARY_PATH: "./build/openvino_genai/:$LD_LIBRARY_PATH"
 
   cpp-prompt_lookup_decoding_lm-ubuntu:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -558,7 +558,7 @@ jobs:
           LD_LIBRARY_PATH: "./build/openvino_genai/:$LD_LIBRARY_PATH"
 
   cpp-Phi-1_5:
-    runs-on: ubuntu-20.04-16-cores
+    runs-on: ubuntu-22.04-16-cores
     defaults:
       run:
         shell: bash
@@ -616,7 +616,7 @@ jobs:
           PYTHONPATH: "./build"
 
   cpp-greedy_causal_lm-redpajama-3b-chat:
-    runs-on: ubuntu-20.04-8-cores
+    runs-on: ubuntu-22.04-8-cores
     defaults:
       run:
         shell: bash
@@ -909,7 +909,7 @@ jobs:
         timeout-minutes: 4
 
   cpp-continuous-batching-ubuntu:
-    runs-on: ubuntu-20.04-8-cores
+    runs-on: ubuntu-22.04-8-cores
     defaults:
       run:
         shell: bash

.github/workflows/job_vlm_sample_llava.yml

@@ -11,7 +11,7 @@ on:
         type: string
 
 env:
-  l_u22_ov_link: https://storage.openvinotoolkit.org/repositories/openvino/packages/nightly/2025.1.0-17911-83c047443de/l_openvino_toolkit_ubuntu22_2025.1.0.dev20250116_x86_64.tgz
+  l_u22_ov_link: https://storage.openvinotoolkit.org/repositories/openvino/packages/nightly/2025.1.0-18141-b0a120c9684/openvino_toolkit_ubuntu22_2025.1.0.dev20250211_x86_64.tgz
 
 jobs:
   visual_language_chat_sample-ubuntu-llava:

.github/workflows/lcm_dreamshaper_cpp.yml

@@ -18,16 +18,16 @@ concurrency:
 
 env:
   PYTHON_VERSION: '3.9'
-  LINUX_OV_ARCHIVE_URL: https://storage.openvinotoolkit.org/repositories/openvino/packages/nightly/2025.1.0-17911-83c047443de/l_openvino_toolkit_ubuntu22_2025.1.0.dev20250116_x86_64.tgz
-  WINDOWS_OV_ARCHIVE_URL: https://storage.openvinotoolkit.org/repositories/openvino/packages/nightly/2025.1.0-17911-83c047443de/w_openvino_toolkit_windows_2025.1.0.dev20250116_x86_64.zip
+  LINUX_OV_ARCHIVE_URL: https://storage.openvinotoolkit.org/repositories/openvino/packages/nightly/2025.1.0-18141-b0a120c9684/openvino_toolkit_ubuntu22_2025.1.0.dev20250211_x86_64.tgz
+  WINDOWS_OV_ARCHIVE_URL: https://storage.openvinotoolkit.org/repositories/openvino/packages/nightly/2025.1.0-18141-b0a120c9684/openvino_toolkit_windows_2025.1.0.dev20250211_x86_64.zip
   OV_INSTALL_DIR: ${{ github.workspace }}/ov
 
 jobs:
   lcm_dreamshaper_v7_cpp-linux:
     runs-on: ubuntu-22.04
     defaults:
       run:
-        shell: bash -l {0}
+        shell: bash
     env:
       build_dir: ${{ github.workspace }}//build
     steps:

.github/workflows/mac.yml

@@ -1,4 +1,4 @@
-name: macOS (12, Python 3.10)
+name: macOS (13, Python 3.10)
 on:
   workflow_dispatch:
   pull_request:
@@ -29,7 +29,7 @@ jobs:
     defaults:
       run:
         shell: bash
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
     steps:
       - name: Download OpenVINO build
         id: openvino_download
@@ -58,7 +58,7 @@ jobs:
     defaults:
       run:
         shell: bash
-    runs-on: 'macos-12-large'
+    runs-on: 'macos-13-large'
     env:
       MACOSX_DEPLOYMENT_TARGET: '10.15'
       CMAKE_BUILD_TYPE: 'Release'

.github/workflows/stable_diffusion_1_5_cpp.yml

@@ -78,7 +78,7 @@ jobs:
     needs: [ openvino_download_linux ]
     defaults:
       run:
-        shell: bash -l {0}
+        shell: bash
     env:
       build_dir: ${{ github.workspace }}//build
       SRC_DIR: ${{ github.workspace }}

.github/workflows/windows.yml

@@ -29,7 +29,7 @@ jobs:
     defaults:
       run:
         shell: bash
-    runs-on: ubuntu-20.04
+    runs-on: ubuntu-22.04
 
     steps:
       - name: Download OpenVINO build

src/python/openvino_genai/py_openvino_genai.pyi

@@ -1535,7 +1535,7 @@ class StreamerBase:
         """
         Put is called every time new token is decoded. Returns a bool flag to indicate whether generation should be stopped, if return true generation stops
         """
-    def write(self, token: int) -> ...:
+    def write(self, token: int) -> StreamingStatus:
         """
         Write is called every time new token is decoded. Returns a StreamingStatus flag to indicate whether generation should be stopped or cancelled
         """

src/python/py_openvino_genai.cpp

@@ -122,17 +122,20 @@ PYBIND11_MODULE(py_openvino_genai, m) {
         .def_readonly("scores", &EncodedResults::scores)
         .def_readonly("perf_metrics", &EncodedResults::perf_metrics);
 
-    py::class_<StreamerBase, ConstructableStreamer, std::shared_ptr<StreamerBase>>(m, "StreamerBase", streamer_base_docstring)  // Change the holder form unique_ptr to shared_ptr
-        .def(py::init<>())
-        .def("put", &StreamerBase::put, "Put is called every time new token is decoded. Returns a bool flag to indicate whether generation should be stopped, if return true generation stops", py::arg("token"))
-        .def("write", &StreamerBase::write, "Write is called every time new token is decoded. Returns a StreamingStatus flag to indicate whether generation should be stopped or cancelled", py::arg("token"))
-        .def("end", &StreamerBase::end, "End is called at the end of generation. It can be used to flush cache if your own streamer has one");
-
     py::enum_<ov::genai::StreamingStatus>(m, "StreamingStatus")
         .value("RUNNING", ov::genai::StreamingStatus::RUNNING)
         .value("CANCEL", ov::genai::StreamingStatus::CANCEL)
         .value("STOP", ov::genai::StreamingStatus::STOP);
 
+    auto streamer = py::class_<StreamerBase, ConstructableStreamer, std::shared_ptr<StreamerBase>>(m, "StreamerBase", streamer_base_docstring)  // Change the holder form unique_ptr to shared_ptr
+        .def(py::init<>())
+        .def("write", &StreamerBase::write, "Write is called every time new token is decoded. Returns a StreamingStatus flag to indicate whether generation should be stopped or cancelled", py::arg("token"))
+        .def("end", &StreamerBase::end, "End is called at the end of generation. It can be used to flush cache if your own streamer has one");
+
+    OPENVINO_SUPPRESS_DEPRECATED_START
+    streamer.def("put", &StreamerBase::put, "Put is called every time new token is decoded. Returns a bool flag to indicate whether generation should be stopped, if return true generation stops", py::arg("token"));
+    OPENVINO_SUPPRESS_DEPRECATED_END
+
     init_tokenizer(m);
     init_lora_adapter(m);
     init_generation_config(m);

tests/python_tests/test_llm_pipeline.py

@@ -172,7 +172,7 @@ def test_chat_scenario_several_chats_in_series():
 
     generation_config_kwargs, _ = chat_intpus[0]
     ov_generation_config = GenerationConfig(**generation_config_kwargs)
-    hf_generation_config = convert_to_hf(opt_model.generation_config, ov_generation_config)
+    hf_generation_config = generation_config_to_hf(opt_model.generation_config, ov_generation_config)
 
     for i in range(2):
         chat_history_hf = []