diff --git a/docs_new/docs/hardware-platforms/ascend-npus/ascend_npu_support_features.mdx b/docs_new/docs/hardware-platforms/ascend-npus/ascend_npu_support_features.mdx index 941cd5c55..75f6e7e94 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/ascend_npu_support_features.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/ascend_npu_support_features.mdx @@ -1241,7 +1241,7 @@ click [Server Arguments](../../advanced_features/server_arguments). `--lora-use-virtual-experts` `False` bool flag
(set to enable) - A2, A3 + Special for GPU `--lora-strict-loading` @@ -1638,7 +1638,7 @@ click [Server Arguments](../../advanced_features/server_arguments). `--deepep-dispatcher-output-dtype` `auto` - `auto`,
`bf16`,
`fp8`,
`int8`,
`nvfp4` + `auto`,
`bf16`,
`fp8`,
`int8` A2, A3 diff --git a/docs_new/docs/hardware-platforms/ascend-npus/model-tutorials/glm_5_2.mdx b/docs_new/docs/hardware-platforms/ascend-npus/model-tutorials/glm_5_2.mdx index 5fb9a9577..b3a6852e5 100644 --- a/docs_new/docs/hardware-platforms/ascend-npus/model-tutorials/glm_5_2.mdx +++ b/docs_new/docs/hardware-platforms/ascend-npus/model-tutorials/glm_5_2.mdx @@ -332,6 +332,7 @@ export GLOO_SOCKET_IFNAME=lo export DEEPEP_NORMAL_LONG_SEQ_ROUND=72 export DEEPEP_NORMAL_LONG_SEQ_PER_ROUND_TOKENS=1024 export DEEPEP_NORMAL_COMBINE_ENABLE_LONG_SEQ=1 +export DEEP_NORMAL_MODE_USE_INT8_QUANT=1 IPS=('' '')