fix dependency issue with --load_quantized_model_with_autoawq

Makefile

@@ -107,7 +107,7 @@ slow_tests_diffusers: test_installs
 
 # Run text-generation non-regression tests
 slow_tests_text_generation_example: test_installs
-	python -m pip install triton==3.1.0 autoawq
+	python -m pip install -r requirements_awq.txt
 	BUILD_CUDA_EXT=0 python -m pip install -vvv --no-build-isolation git+https://github.com/HabanaAI/AutoGPTQ.git
 	python -m pip install git+https://github.com/HabanaAI/[email protected]
 	python -m pytest tests/test_text_generation_example.py tests/test_encoder_decoder.py -v -s --token $(TOKEN)

examples/text-generation/README.md

@@ -735,7 +735,7 @@ Currently, this support is limited to UINT4 inference of pre-quantized models on
 
 Please run the following command to install AutoAWQ:
 ```bash
-pip install triton==3.1.0 autoawq
+pip install -r requirements_awq.txt
 ```
 
 You can run a *UINT4 weight quantized* model using AutoAWQ by including the argument `--load_quantized_model_with_autoawq`.

examples/text-generation/requirements_awq.txt

@@ -0,0 +1,2 @@
+triton==3.1.0
+autoawq