Commits · f38ca71cab46332e313a8cc3b1ef09a20548e1e6 · jerrrrry / infinicore

04 Sep, 2025 1 commit
- issue/423: improve the precision of the torch implementation of rms_norm · f38ca71c
  PanZezhong1725 authored Sep 04, 2025
  
  f38ca71c
03 Sep, 2025 15 commits
- Merge pull request #417 from InfiniTensor/issue/416 · c7373fee
  zhangyue authored Sep 03, 2025
```
issue/416: p800 rearrange kernel
```
  c7373fee
- issue/416: 整理文件结构 · f0c357b7
  zhangyue authored Sep 02, 2025
  
  f0c357b7
- issue/416: p800 rearrange kernel · 8c798185
  zhangyue authored Sep 02, 2025
  
  8c798185
- Merge pull request #358 from InfiniTensor/issue/342 · 0e1c5585
  zhangyue authored Sep 03, 2025
```
issue/342: 昆仑芯P800上random_sample算子
```
  0e1c5585
- issue/342: delete to · 1cadb2a1
  xgqdut2016 authored Sep 03, 2025
  
  1cadb2a1
- issue/423: improve the precision of the torch implementation of rms_norm · 612defae
  Ziminli authored Sep 03, 2025
  
  612defae
- issue/342: modified loadsm · 0ecbe1d5
  xgqdut2016 authored Sep 03, 2025
  
  0ecbe1d5
- issue/342: topk and softmaxsum · 79b3acc3
  xgqdut2016 authored Sep 03, 2025
  
  79b3acc3
- issue/342: modified workspace · 711fecf4
  xgqdut2016 authored Sep 02, 2025
  
  711fecf4
- issue/342: success random_sample all · d741ee7d
  xgqdut2016 authored Sep 02, 2025
  
  d741ee7d
- issue/342: F16 success but BF16 failed · c5bc6628
  xgqdut2016 authored Sep 01, 2025
  
  c5bc6628
- issue/342: success kunlunP800 random_sample · a4b897d9
  xgqdut2016 authored Aug 12, 2025
  
  a4b897d9
- Merge pull request #422 from InfiniTensor/issue/421 · 19d60bf8
  zhangyue authored Sep 03, 2025
```
issue/421: 适配 rmsnorm 测例修改，支持 bf16 和 f16数据类型 weights
```
  19d60bf8
- issue/421: 适配 rmsnorm 测例修改，支持 bf16 和 f16数据类型 weights · 1048c1bc
  zhangyue authored Sep 03, 2025
  
  1048c1bc
- Merge pull request #419 from InfiniTensor/issue/418 · c0d1b0d0
  zhangyue authored Sep 03, 2025
```
issue/418: 解决 p800 上手写算子引用 sm 上指针的报错问题
```
  c0d1b0d0
02 Sep, 2025 2 commits
- [T2-2-3] blkmjsian · 9ad23fad
  blkmjsian authored Sep 02, 2025
```
- dequantize awq
- rope v2
```
  9ad23fad
- issue/418: 解决 p800 上手写算子引用 sm 上指针的报错问题 · 20488ee7
  zhangyue authored Sep 02, 2025
  
  20488ee7
29 Aug, 2025 1 commit
- issue/411: adjust mudnn handle use to prevent mudnn handle mismatching · b3170335
  PanZezhong1725 authored Aug 29, 2025
  
  b3170335
28 Aug, 2025 2 commits
- fix: implement mudnn handle use to prevent mudnn handle mismatching · f34d4e3b
  zhushuang authored Aug 28, 2025
  
  f34d4e3b
- Merge pull request #409 from InfiniTensor/issue/340 · 1d064392
  zhangyue authored Aug 28, 2025
```
Issue/340 接入昆仑芯 XBLAS
```
  1d064392
27 Aug, 2025 7 commits
- issue/340: xblas · b92ecc31
  zhangyue authored Aug 27, 2025
  
  b92ecc31
- issue/406: 修复 isMergable 的 stride 检查逻辑 (#408) · e221916d
  PanZezhong1725 authored Aug 27, 2025
  
  e221916d
- issue/260: 摩尔平台 causal_softmax 算子开发 · f50c2a40
  spike-zhu authored Aug 27, 2025
  
  f50c2a40
- issue/251: 摩尔平台 rope 算子开发 · 9c87dbb1
  spike-zhu authored Aug 27, 2025
  
  9c87dbb1
- issue/294：摩尔通信库接入 · f2bb97ad
  spike-zhu authored Aug 27, 2025
  
  f2bb97ad
- Issue/262: 摩尔平台 random_sample · d3d6fee0
  spike-zhu authored Aug 27, 2025
  
  d3d6fee0
- issue/406: 修复 isMergable 的 stride 检查逻辑 · ad4a2be2
  MaYuhang authored Aug 27, 2025
  
  ad4a2be2
26 Aug, 2025 5 commits
- issue/340: kunlun cublas gemm · a9acf208
  xgqdut2016 authored Aug 05, 2025
  
  a9acf208
- Merge pull request #405 from InfiniTensor/issue/404 · cb06c721
  zhangyue authored Aug 26, 2025
```
issue/404: kunlun_common.h 和 kunlun_handle.h 解耦合
```
  cb06c721
- issue/404 修改 #define · 376065bc
  zhangyue authored Aug 26, 2025
  
  376065bc
- issue/404: 删除无用注释 · b95d7c00
  zhangyue authored Aug 26, 2025
  
  b95d7c00
- issue/404: kunlun_common.h 和 kunlun_handle.h 解耦合 · 2fe7052f
  zhangyue authored Aug 26, 2025
  
  2fe7052f
25 Aug, 2025 4 commits
- Merge pull request #391 from InfiniTensor/issue/390 · 97a3a84e
  zhangyue authored Aug 25, 2025
```
issue/390: kunlun p800 causal softmax
```
  97a3a84e
- issue/390: kunlun p800 causal softmax · c5167eb7
  zhangyue authored Aug 22, 2025
  
  c5167eb7
- Merge pull request #400 from InfiniTensor/issue/388-rmsnorm3d-p800 · c35920e2
  zhangyue authored Aug 25, 2025
```
issue390: p800 rmsnorm 3d
```
  c35920e2
- p800 rmsnorm 3d · e6dd5604
  zhangyue authored Aug 25, 2025
  
  e6dd5604
22 Aug, 2025 3 commits
- Merge pull request #284 from InfiniTensor/issue/270 · c9f5f2ec
  PanZezhong1725 authored Aug 22, 2025
```
Issue/270: 摩尔平台 reareange 算子开发
```
  c9f5f2ec
- issue/388: Support 3D Cases for RMS Norm · dc8ddd58
  Ziminli authored Aug 22, 2025
  
  dc8ddd58
- feat: add rearrange in moore gpu, with test pass · d1964ef4
  zhushuang authored Jun 27, 2025
  
  d1964ef4