Commits · ac4aae48f5d3a70c19ce48a4c7c6081c48096d19 · jerrrrry / infinicore

01 Dec, 2025 11 commits
- Merge branch 'main' into dev_topkrouter · ac4aae48
  Shijie authored Dec 01, 2025
  
  ac4aae48
- issue/634: InfiniCore 支持InfiniLM Llama模型适配 (#668) · 2f3f4076
  Ceng authored Dec 01, 2025
```
* issue/634: InfiniCore 支持InfiniLM Llama模型适配
Signed-off-by: Ceng23333 <441651826@qq.com>

* .
Signed-off-by: Ceng23333 <441651826@qq.com>

---------
Signed-off-by: Ceng23333 <441651826@qq.com>
```
  2f3f4076
- Merge pull request #681 from pengcheng888/issue/679_1 · 1bafd1a6
  pengcheng888 authored Dec 01, 2025
```
issue/679 - 减少创建和使用infini.tensor对象过程中的属性调用次数，以降低耗时
```
  1bafd1a6
- Merge pull request #677 from InfiniTensor/issue/676 · 1887c3f1
  zhangyue authored Dec 01, 2025
```
Issue/676 昆仑芯 topkrouter
```
  1887c3f1
- Merge pull request #686 from InfiniTensor/issue/684 · e290ff22
  qinyiqun authored Dec 01, 2025
```
issue/684: QY机器添加通信库编译选项
```
  e290ff22
- Issue/672 a (#673) · e16b6dbc
  thatPepe authored Dec 01, 2025
```
* issue/672 - fixed non-contiguous operations in the test framework
```
  e16b6dbc
- issue/684: QY机器添加通信库编译选项 · b17a89df
  xgqdut2016 authored Dec 01, 2025
  
  b17a89df
- Merge pull request #683 from pengcheng888/issue/680_mm · 79e6883b
  pengcheng888 authored Dec 01, 2025
```
issue/680-为mulmat添加alpha参数算子
```
  79e6883b
- issue/680-为mulmat添加alpha参数算子 · 4916ff8a
  pengcheng888 authored Dec 01, 2025
  
  4916ff8a
- issue/676: fix format · 56ec08ae
  zhangyue authored Dec 01, 2025
  
  56ec08ae
- issue/679 - 减少创建和使用infini.tensor过程中的属性调用次数，以减少耗时 · 0c0489c5
  pengcheng888 authored Dec 01, 2025
  
  0c0489c5
29 Nov, 2025 1 commit
- issue/563 Add metax support for topkrouter · a15aa367
  Zhao Shijie authored Nov 28, 2025
  
  a15aa367
28 Nov, 2025 6 commits
- issue/676: fixing · 6cf95c95
  zhangyue authored Nov 28, 2025
  
  6cf95c95
- issue/676: fix format · ce10d777
  zhangyue authored Nov 28, 2025
  
  ce10d777
- issue/676 format · 74aeb4f4
  zhangyue authored Nov 28, 2025
  
  74aeb4f4
- issue/676: format · c1af9783
  zhangyue authored Nov 28, 2025
  
  c1af9783
- issue/676: kunlun topkrouter · 5584035d
  zhangyue authored Nov 28, 2025
  
  5584035d
- issue/614: fix device class in python · 5f0f80d6
  zhangyue authored Nov 18, 2025
  
  5f0f80d6
26 Nov, 2025 2 commits
- Merge pull request #671 from gongchensu/feature/metax_op_fixed · cad2d45a
  PanZezhong1725 authored Nov 26, 2025
```
fix: correct macro for mccub/hccub conditional compilation.
```
  cad2d45a
- Issue/670 - fix: correct macro for mccub/hccub conditional compilation. · 904a9254
  zhuyue authored Nov 26, 2025
  
  904a9254
25 Nov, 2025 2 commits
- Merge pull request #667 from InfiniTensor/issue/666 · 62797984
  PanZezhong1725 authored Nov 25, 2025
```
issue/666 - Standardized test imports
```
  62797984
- Merge pull request #613 from InfiniTensor/issue/612 · 5e85a4d8
  PanZezhong1725 authored Nov 25, 2025
```
issue/612 - differentiate mempcy sync/async
```
  5e85a4d8
24 Nov, 2025 1 commit
- issue/666 - Standardized test imports · 61a7dc0e
  wooway777 authored Nov 24, 2025
  
  61a7dc0e
22 Nov, 2025 13 commits
- Merge pull request #663 from gongchensu/feature/moore_adapt · 406c9668
  PanZezhong1725 authored Nov 22, 2025
```
Issue/658 - Update test tolerances and remove device-specific dtype f…
```
  406c9668
- Issue/658 - Update test tolerances and remove device-specific dtype filters. · e00e65e2
  zhuyue authored Nov 22, 2025
  
  e00e65e2
- Merge pull request #600 from InfiniTensor/issue/599 · 3aeee034
  PanZezhong1725 authored Nov 22, 2025
```
将 `README.md` 中的 `pip install . -e` 改为 `pip install -e .`
```
  3aeee034
- Merge pull request #662 from gongchensu/feature/moore_adapt · 35bf00bb
  PanZezhong1725 authored Nov 22, 2025
```
Issue/661 - Support run.py --bench
```
  35bf00bb
- Issue/661 - Support run.py --bench · ed95eb04
  zhuyue authored Nov 22, 2025
  
  ed95eb04
- Merge pull request #659 from gongchensu/feature/moore_adapt · 20a45b60
  gongchensu authored Nov 22, 2025
```
Feature/moore adapt
```
  20a45b60
- Issue/658: Skip non-contiguous tensor tests on Moore platform · c87a2ef4
  zhuyue authored Nov 22, 2025
  
  c87a2ef4
- Issue/658 - Comment out unsupported float16/bfloat16 tests in isin, mse_loss, and pad. · 314d4a65
  zhuyue authored Nov 22, 2025
  
  314d4a65
- Issue/658 - Add Moore platform support for add, mul, and silu operations · 33d0f769
  zhuyue authored Nov 22, 2025
```
- Implement Moore backend for add, mul, and silu elementwise operations
- Filter unsupported dtypes (BF16, F64) for Moore platform in tests
```
  33d0f769
- Merge pull request #657 from InfiniTensor/issue/656 · dfd1341e
  PanZezhong1725 authored Nov 22, 2025
```
issue/656 - updated readme
```
  dfd1341e
- issue/656 - updated readme · 5e2b003c
  wooway777 authored Nov 22, 2025
  
  5e2b003c
- modify stack operator test · 77ad94a9
  Your Name authored Nov 21, 2025
  
  77ad94a9
- Add max_unpool 1D/2D/3D and stack operator test cases · fe164657
  Your Name authored Nov 20, 2025
  
  fe164657
21 Nov, 2025 4 commits
- Merge pull request #651 from gongchensu/feature/metax_op_fixed · 0298925d
  gongchensu authored Nov 21, 2025
```
Issue/648 - Remove metax device support for layer_norm, softmax, and …
```
  0298925d
- Issue/654 - Update CUB API usage for CUDA 12.9+ compatibility · 10572e55
  zhuyue authored Nov 21, 2025
  
  10572e55
- Merge pull request #655 from gongchensu/feature/fix_cuda_version · 77d4ac50
  PanZezhong1725 authored Nov 21, 2025
```
Issue/654 - Fix CUDA 13.0 compatibility issues
```
  77d4ac50
- Issue/654 - Update CUB API usage for CUDA 12.9+ compatibility · 9a9f0982
  zhuyue authored Nov 21, 2025
  
  9a9f0982