[FlashInver v0.6.7] Integrate flashinfer_trtllm mxfp8 gemm (#21576)
This commit is contained in:
@@ -131,5 +131,10 @@ class TestMXFP8GemmFlashinferTrtllm(MXFP8GemmBase, unittest.TestCase):
|
||||
backend = "flashinfer_trtllm"
|
||||
|
||||
|
||||
@unittest.skipIf(get_device_sm() < 100, "Test requires CUDA SM 100 or higher")
|
||||
class TestMXFP8GemmFlashinferCutlass(MXFP8GemmBase, unittest.TestCase):
|
||||
backend = "flashinfer_cutlass"
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
unittest.main()
|
||||
|
||||
Reference in New Issue
Block a user