Default Branch

08c5d4ea14 · Update README.md · Updated 2024-08-20 22:10:21 +08:00

Branches

9a5c848616 · modify CMake_version · Updated 2024-08-23 16:03:45 +08:00

3
5

fc09933fc3 · ascend plugin sub kernel 支持新的数据形状 · Updated 2024-08-22 16:15:05 +08:00

16
61

a5b20f191f · fix · Updated 2024-08-22 10:51:54 +08:00

2
11

a21c76f619 · fix domestic runtime · Updated 2024-08-19 17:13:26 +08:00

3
1

9838d9370d · Add conv3d op && change ffi · Updated 2024-08-16 16:18:43 +08:00

16
58

91f1707eb2 · 添加 MLU 平台分布式验收脚本 (#223) · Updated 2024-08-08 18:36:57 +08:00

19
21

3c2a991db9 · fix(matmul): fix the data type conversion function for matmul. · Updated 2024-08-07 14:28:23 +08:00

5
2

c00065be8f · modified format · Updated 2024-08-02 10:45:01 +08:00

5
2

264901b5e3 · pingpong speed · Updated 2024-07-24 16:57:51 +08:00

86
17

589154304e · style · Updated 2024-07-16 11:19:09 +08:00

5
5

7861d38048 · modified format · Updated 2024-07-04 09:40:12 +08:00

7
7

6a0659fab6 · Merge branch 'master' into cuda-transpose · Updated 2024-06-12 11:13:17 +08:00

15
6

3b81100a46 · We handle null alpha when we load onnx model · Updated 2024-06-05 16:06:32 +08:00

27
3

20f651b1d3 · implement instance norm in front · Updated 2024-05-08 17:44:54 +08:00

16
2

b0d030d0de · [fix] fix rope op test failing · Updated 2024-04-23 13:51:10 +08:00

19
19

4a5b9572bb · add test scripts for llama2 and 9G models · Updated 2024-04-10 16:23:02 +08:00

19
17

3b7b5740af · allocate workspace from allocator for kunlun runtime · Updated 2024-04-08 15:48:06 +08:00

20
5

6c4dd7b28b · fix(front): 将stub改为可以接收GraphProto作为输入,消除分布式脚本保存额外的onnx文件, 采用int64作为index输入类型 · Updated 2024-04-07 17:15:40 +08:00

19
1

25a3cedeb0 · add pytorch bench · Updated 2024-03-21 10:27:32 +08:00

24
1

e33131ce5c · fix comment · Updated 2024-01-17 10:57:44 +08:00

41
10