Implement NO_KERNELS flag and update torch requirement (#582)

Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>

setup.py

@@ -33,6 +33,7 @@ def get_kernels_whl_url(
 
 AUTOAWQ_VERSION = "0.2.6"
 PYPI_BUILD = os.getenv("PYPI_BUILD", "0") == "1"
+NO_KERNELS = int(os.getenv("NO_KERNELS", "0"))
 IS_CPU_ONLY = not torch.backends.mps.is_available() and not torch.cuda.is_available()
 
 CUDA_VERSION = os.getenv("CUDA_VERSION", None) or torch.version.cuda
@@ -86,7 +87,7 @@ def get_kernels_whl_url(
 }
 
 requirements = [
-    "torch==2.3.1",
+    "torch>=2.3.1",
     "transformers>=4.35.0",
     "tokenizers>=0.12.1",
     "typing_extensions>=4.8.0",
@@ -107,7 +108,7 @@ def get_kernels_whl_url(
 
 # kernels can be downloaded from pypi for cuda+121 only
 # for everything else, we need to download the wheels from github
-if not KERNELS_INSTALLED and (CUDA_VERSION or ROCM_VERSION):
+if not KERNELS_INSTALLED and (CUDA_VERSION or ROCM_VERSION) and not NO_KERNELS:
     if CUDA_VERSION and CUDA_VERSION.startswith("12"):
         requirements.append("autoawq-kernels")
     elif CUDA_VERSION and CUDA_VERSION.startswith("11") or ROCM_VERSION in ["571"]: