apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Metax] adapt DeepSeek (#4498)

Browse Source
This commit is contained in:
xiaozude
2025-10-24 10:14:53 +08:00
committed by GitHub
parent 8718fa34b2
commit f7069b8057
19 changed files with 1538 additions and 324 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
custom_ops/setup_ops.py
View File

@@ -601,9 +601,16 @@ elif paddle.device.is_compiled_with_custom_device("metax_gpu"):
"gpu_ops/read_data_ipc.cu",
"gpu_ops/dequant_int8.cu",
"gpu_ops/share_external_data.cu",
"gpu_ops/recover_decode_task.cu",
"gpu_ops/noaux_tc.cu",
"gpu_ops/fused_rotary_position_encoding.cu",
"gpu_ops/text_image_gather_scatter.cu",
"gpu_ops/text_image_index_out.cu",
"gpu_ops/get_position_ids_and_mask_encoder_batch.cu",
"gpu_ops/append_attn/mla_cache_kernel.cu",
"gpu_ops/append_attn/get_block_shape_and_split_kv_block.cu",
"gpu_ops/moe/tritonmoe_preprocess.cu",
"gpu_ops/moe/moe_topk_select.cu",
"gpu_ops/recover_decode_task.cu",
"metax_ops/moe_dispatch.cu",
"metax_ops/moe_ffn.cu",
"metax_ops/moe_reduce.cu",