infra: fix models cache GHA (vllm-project#238)

* infra: test model cache

Signed-off-by: yuluo-yx <[email protected]>

* chore: lookup huggingface cache dir

Signed-off-by: yuluo-yx <[email protected]>

* feat: when run test-vllm, get model from openai models api (vllm-project#236)

Signed-off-by: yuluo-yx <[email protected]>

* infra: cache models in test-and-build GHA (vllm-project#237)

Signed-off-by: yuluo-yx <[email protected]>

* chore: lookup huggingface cache dir

Signed-off-by: yuluo-yx <[email protected]>

* fix: only cache models

Signed-off-by: yuluo-yx <[email protected]>

* chore

Signed-off-by: yuluo-yx <[email protected]>

---------

Signed-off-by: yuluo-yx <[email protected]>

Author: yuluo-yx

.github/workflows/test-and-build.yml

@@ -56,7 +56,6 @@ jobs:
       with:
         path: |
           models/
-          ~/.cache/huggingface/
         key: ${{ runner.os }}-models-v1-${{ hashFiles('tools/make/models.mk') }}
         restore-keys: |
           ${{ runner.os }}-models-v1-

tools/make/models.mk

@@ -19,7 +19,7 @@ download-models:
 		hf download LLM-Semantic-Router/pii_classifier_modernbert-base_presidio_token_model --local-dir models/pii_classifier_modernbert-base_presidio_token_model; \
 	fi
 
-	@if [ ! -d "lora_intent_classifier_bert-base-uncased_model" ]; then \
+	@if [ ! -d "models/lora_intent_classifier_bert-base-uncased_model" ]; then \
 		hf download LLM-Semantic-Router/lora_intent_classifier_bert-base-uncased_model --local-dir models/lora_intent_classifier_bert-base-uncased_model; \
 	fi