【NPU】【bugfix】fix server error when mtp unquant (#26389)

Co-authored-by: cen121212 <luochen23@huawei.com> Co-authored-by: Even Zhou <even.y.zhou@outlook.com>
2026-05-30 15:01:19 +03:00
parent 282c46133f
commit b421e60eed
5 changed files with 148 additions and 98 deletions
@@ -60,6 +60,7 @@ class TestAscendDeepEP(CustomTestCase):
            "SGLANG_DEEPEP_NUM_MAX_DISPATCH_TOKENS_PER_RANK": "32",
            "SGLANG_NPU_USE_MLAPO": "1",
            "TRANSFORMERS_VERBOSITY": "error",
+            "DEEP_NORMAL_MODE_USE_INT8_QUANT": "1",
        }
        os.environ.update(cls.extra_envs)