Bump transformers to 5.0.0

[AlexanderDokuchaev]

Changes

Bump transformers to 5.0.0

• apply_chat_template returns BatchFeature instance instead of dict
• Install eval_llm with [hf] to install all requaread packages
• Removed StaticCacheConfig
• Updated WWB fork
• Removed dummy_llama tests for SparsityActivation alogrimth as it falls

Reason for changes

Related tickets

Tests

Test examples - success
Weight compression - success
PTQ-873

[Copilot]

Pull request overview

This PR updates NNCF’s test and example environments to be compatible with transformers==5.0.0, adjusting call sites for API changes (notably apply_chat_template return type) and updating generation static-cache configuration accordingly.

Changes:

• Bump transformers to 5.0.0 across test/example requirements and update related dependencies (e.g., sentence-transformers, tensorflow-io, whowhatbench, lm_eval[hf]).
• Update example code to handle tokenizer.apply_chat_template(...) returning a BatchFeature (extracting input_ids explicitly).
• Remove StaticCacheConfig usage and switch generation cache_config to a dict-based configuration; remove the dummy_llama sparsify-activations unit test helper/testcase.

Reviewed changes

Copilot reviewed 25 out of 25 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
tests/torch/requirements.txt	Pins transformers==5.0.0 and updates sentence-transformers for torch test environment.
tests/torch/function_hook/sparsify_activations/test_algo.py	Removes dummy_llama sparsify-activations algorithm test case.
tests/torch/function_hook/sparsify_activations/helpers.py	Removes dummy_llama_model helper and related transformers import.
tests/post_training/requirements.txt	Pins transformers==5.0.0, bumps tensorflow-io, updates WWB fork commit.
tests/post_training/pipelines/fx_modelling.py	Drops StaticCacheConfig and updates static cache setup for Transformers 5.
tests/openvino/requirements.txt	Pins transformers==5.0.0 for OpenVINO tests.
examples/llm_compression/torch/downstream_qat_with_nls/requirements.txt	Pins transformers==5.0.0 and switches to lm_eval[hf].
examples/llm_compression/torch/distillation_qat_with_lora/requirements.txt	Pins transformers==5.0.0 and switches to lm_eval[hf].
examples/llm_compression/torch_fx/tiny_llama/requirements.txt	Pins transformers==5.0.0 for the torch.fx tiny-llama example.
examples/llm_compression/torch_fx/tiny_llama/modelling.py	Updates static cache configuration to dict and ensures use_cache is enabled.
examples/llm_compression/torch_fx/tiny_llama/main.py	Adjusts chat template tokenization to handle BatchFeature output.
examples/llm_compression/openvino/tiny_llama/requirements.txt	Pins transformers==5.0.0 for OpenVINO tiny-llama example.
examples/llm_compression/openvino/tiny_llama/main.py	Adjusts chat template tokenization to handle BatchFeature output.
examples/llm_compression/openvino/tiny_llama_synthetic_data/requirements.txt	Pins transformers==5.0.0 for synthetic-data example.
examples/llm_compression/openvino/tiny_llama_find_hyperparams/requirements.txt	Pins transformers==5.0.0 and updates WWB fork commit.
examples/llm_compression/openvino/smollm2_360m_fp8/requirements.txt	Pins transformers==5.0.0 for FP8 example.
examples/llm_compression/openvino/smollm2_360m_fp8/main.py	Adjusts chat template tokenization to handle BatchFeature output.
examples/llm_compression/openvino/smollm2_360m_codebook/requirements.txt	Pins transformers==5.0.0 for codebook example.
examples/llm_compression/openvino/smollm2_360m_codebook/main.py	Adjusts chat template tokenization to handle BatchFeature output.
examples/llm_compression/openvino/smollm2_360m_adaptive_codebook/requirements.txt	Pins transformers==5.0.0 for adaptive codebook example.
examples/llm_compression/openvino/smollm2_360m_adaptive_codebook/main.py	Adjusts chat template tokenization to handle BatchFeature output.
examples/llm_compression/onnx/tiny_llama/requirements.txt	Pins transformers==5.0.0 for ONNX tiny-llama example.
examples/llm_compression/onnx/tiny_llama/main.py	Adjusts chat template tokenization to handle BatchFeature output.
examples/llm_compression/onnx/tiny_llama_scale_estimation/requirements.txt	Pins transformers==5.0.0 for scale-estimation example.
examples/llm_compression/onnx/tiny_llama_scale_estimation/main.py	Adjusts chat template tokenization to handle BatchFeature output.