transformers v5 upgrade (#3272)

* Prepare for transformers v5 upgrade

* fix hf cli

* update for hf hub changes

* fix tokenizer apply_chat_template args

* remap include_tokens_per_second

* fix tps

* handle migration for warmup

* use latest hf hub

* Fix scan -> ls

* fix import

* fix for renaming of mistral common tokenizer -> backend

* update for fixed tokenziation for llama

* Skip phi35 tests for now

* remove mistral patch fixed upstream in huggingface/transformers#41439

* use namespacing for patch

* don't rely on sdist for e2e tests for now

* run modal ci without waiting too

* Fix dep for ci

* fix imports

* Fix fp8 check

* fsdp2 fixes

* fix version handling

* update fsdp version tests for new v5 behavior

* Fail multigpu tests after 3 failures

* skip known v5 broken tests for now and cleanup

* bump deps

* unmark skipped test

* re-enable test_fsdp_qlora_prequant_packed test

* increase multigpu ci timeout

* skip broken gemma3 test

* reduce timout back to original 120min now that the hanging test is skipped

* fix for un-necessary collator for pretraining with bsz=1

* fix: safe_serialization deprecated in transformers v5 rc01 (#3318)

* torch_dtype deprecated

* load model in float32 for consistency with tests

* revert some test fixtures back

* use hf cache ls instead of scan

* don't strip fsdp_version

more fdsp_Version fixes for v5
fix version in fsdp_config
fix aliasing
fix fsdp_version check
check fsdp_version is 2 in both places

* Transformers v5 rc2 (#3347)

* bump dep

* use latest fbgemm, grab model config as part of fixture, un-skip test

* import AutoConfig

* don't need more problematic autoconfig when specifying config.json manually

* add fixtures for argilla ultrafeedback datasets

* download phi4-reasoning

* fix arg

* update tests for phi fast tokenizer changes

* use explicit model types for gemma3

---------

Co-authored-by: Wing Lian <wing@axolotl.ai>

* fix: AutoModelForVision2Seq -> AutoModelForImageTextToText

* chore: remove duplicate

* fix: attempt fix gemma3 text mode

* chore: lint

* ga release of v5

* need property setter for name_or_path for mistral tokenizer

* vllm not compatible with transformers v5

* setter for chat_template w mistral too

---------

Co-authored-by: NanoCode012 <nano@axolotl.ai>
Co-authored-by: salman <salman.mohammadi@outlook.com>

tests/conftest.py

@@ -83,6 +83,12 @@ def download_smollm2_135m_model():
     snapshot_download_w_retry("HuggingFaceTB/SmolLM2-135M", repo_type="model")
 
 
+@pytest.fixture(scope="session", autouse=True)
+def download_smollm2_135m_instruct_model():
+    # download the model
+    snapshot_download_w_retry("HuggingFaceTB/SmolLM2-135M-Instruct", repo_type="model")
+
+
 @pytest.fixture(scope="session", autouse=True)
 def download_smollm2_135m_gptq_model():
     # download the model
@@ -143,12 +149,20 @@ def download_argilla_distilabel_intel_orca_dpo_dataset():
     )
 
 
-# @pytest.fixture(scope="session", autouse=True)
-# def download_argilla_ultrafeedback_binarized_preferences_cleaned_dataset():
-#     # download the dataset
-#     snapshot_download_w_retry(
-#         "argilla/ultrafeedback-binarized-preferences-cleaned", repo_type="dataset"
-#     )
+@pytest.fixture(scope="session", autouse=True)
+def download_argilla_ultrafeedback_binarized_preferences_cleaned_dataset():
+    # download the dataset
+    snapshot_download_w_retry(
+        "argilla/ultrafeedback-binarized-preferences-cleaned", repo_type="dataset"
+    )
+
+
+@pytest.fixture(scope="session", autouse=True)
+def download_argilla_ultrafeedback_binarized_preferences_cleaned_kto_dataset():
+    # download the dataset
+    snapshot_download_w_retry(
+        "argilla/ultrafeedback-binarized-preferences-cleaned-kto", repo_type="dataset"
+    )
 
 
 # @pytest.fixture(scope="session", autouse=True)
@@ -251,7 +265,9 @@ def download_llama_1b_model_fixture():
 def download_llama3_8b_model_fixture():
     # download the tokenizer only
     snapshot_download_w_retry(
-        "NousResearch/Meta-Llama-3-8B", repo_type="model", allow_patterns=["*token*"]
+        "NousResearch/Meta-Llama-3-8B",
+        repo_type="model",
+        allow_patterns=["*token*", "config.json"],
     )
 
 
@@ -261,7 +277,7 @@ def download_llama3_8b_instruct_model_fixture():
     snapshot_download_w_retry(
         "NousResearch/Meta-Llama-3-8B-Instruct",
         repo_type="model",
-        allow_patterns=["*token*"],
+        allow_patterns=["*token*", "config.json"],
     )
 
 
@@ -269,7 +285,19 @@ def download_llama3_8b_instruct_model_fixture():
 def download_phi_35_mini_model_fixture():
     # download the tokenizer only
     snapshot_download_w_retry(
-        "microsoft/Phi-3.5-mini-instruct", repo_type="model", allow_patterns=["*token*"]
+        "microsoft/Phi-3.5-mini-instruct",
+        repo_type="model",
+        allow_patterns=["*token*", "config.json"],
+    )
+
+
+@pytest.fixture(scope="session", autouse=True)
+def download_phi_4_reasoning_model_fixture():
+    # download the tokenizer only
+    snapshot_download_w_retry(
+        "microsoft/Phi-4-reasoning",
+        repo_type="model",
+        allow_patterns=["*token*", "config.json"],
     )
 
 
@@ -279,7 +307,7 @@ def download_phi_3_medium_model_fixture():
     snapshot_download_w_retry(
         "microsoft/Phi-3-medium-128k-instruct",
         repo_type="model",
-        allow_patterns=["*token*"],
+        allow_patterns=["*token*", "config.json"],
     )
 
 
@@ -562,6 +590,8 @@ def test_load_fixtures(
     download_mhenrichsen_alpaca_2k_dataset,
     download_mhenrichsen_alpaca_2k_w_revision_dataset,
     download_mlabonne_finetome_100k_dataset,
+    download_argilla_ultrafeedback_binarized_preferences_cleaned_dataset,
+    download_argilla_ultrafeedback_binarized_preferences_cleaned_kto_dataset,
     download_argilla_distilabel_capybara_dpo_7k_binarized_dataset,
     download_arcee_ai_distilabel_intel_orca_dpo_pairs_dataset,
     download_argilla_dpo_pairs_dataset,
@@ -573,6 +603,7 @@ def test_load_fixtures(
     download_llama3_8b_instruct_model_fixture,
     download_phi_35_mini_model_fixture,
     download_phi_3_medium_model_fixture,
+    download_phi_4_reasoning_model_fixture,
     download_mistral_7b_model_fixture,
     download_gemma_2b_model_fixture,
     download_gemma2_9b_model_fixture,