Commits · 7ecee3432110bae563c8756a66b54e5f08dc777d · OpenDAS / vllm_cscc

01 Aug, 2024 1 commit
- [Kernel][RFC] Refactor the punica kernel based on Triton (#5036) · 7ecee343
  Jee Jee Li authored Aug 01, 2024
  
  7ecee343
27 Jul, 2024 1 commit
- [ROCm] Upgrade PyTorch nightly version (#6845) · ced36cd8
  Woosuk Kwon authored Jul 26, 2024
  
  ced36cd8
26 Jul, 2024 1 commit
- [Build/CI][ROCm] Minor simplification to Dockerfile.rocm (#6811) · 71734f1b
  Woosuk Kwon authored Jul 26, 2024
  
  71734f1b
20 Jul, 2024 1 commit
- [Bugfix][CI/Build][Hardware][AMD] Fix AMD tests, add HF cache, update CK FA,... · 06d6c5fe
  Matt Wong authored Jul 20, 2024
```
[Bugfix][CI/Build][Hardware][AMD] Fix AMD tests, add HF cache, update CK FA, add partially supported model notes (#6543)
```
  06d6c5fe
17 Jul, 2024 1 commit
- [ROCm] Cleanup Dockerfile and remove outdated patch (#6482) · 10383887
  Hongxia Yang authored Jul 17, 2024
  
  10383887
12 Jul, 2024 1 commit
- [ROCm][AMD] unify CUDA_VISIBLE_DEVICES usage in cuda/rocm (#6352) · b6c16cf8
  Hongxia Yang authored Jul 12, 2024
  
  b6c16cf8
29 Jun, 2024 1 commit
- [Bugfix][CI/Build][Hardware][AMD] Install matching torchvision to fix AMD tests (#5949) · 9def1066
  Matt Wong authored Jun 29, 2024
  
  9def1066
25 Jun, 2024 1 commit
- [Hardware][AMD][CI/Build][Doc] Upgrade to ROCm 6.1, Dockerfile improvements, test fixes (#5422) · dd793d1d
  Matt Wong authored Jun 25, 2024
  
  dd793d1d
19 Jun, 2024 1 commit
- [Bugfix][CI/Build][AMD][ROCm]Fixed the cmake build bug which generate garbage... · f758aed0
  Hongxia Yang authored Jun 19, 2024
```
[Bugfix][CI/Build][AMD][ROCm]Fixed the cmake build bug which generate garbage on certain devices (#5641)
```
  f758aed0
15 Jun, 2024 1 commit
- Add ccache to amd (#5555) · bd7efe95
  Simon Mo authored Jun 14, 2024
  
  bd7efe95
09 Jun, 2024 1 commit
- [Kernel][Misc] Use TORCH_LIBRARY instead of PYBIND11_MODULE for custom ops (#5047) · 5467ac31
  bnellnm authored Jun 09, 2024
  
  5467ac31
02 Jun, 2024 1 commit
- [Kernel][ROCm][AMD] enable fused topk_softmax kernel for moe layer (#4927) · a66cf40b
  Divakar Verma authored Jun 02, 2024
```
This PR enables the fused topk_softmax kernel used in moe layer for HIP
```
  a66cf40b
20 May, 2024 1 commit
- [Build/CI] Enabling AMD Entrypoints Test (#4834) · 943e72ca
  Alexei-V-Ivanov-AMD authored May 20, 2024
```
Co-authored-by: Alexey Kondratiev <alexey.kondratiev@amd.com>
```
  943e72ca
09 May, 2024 1 commit
- [ROCm] Add support for Punica kernels on AMD GPUs (#3140) · ff5abcd7
  kliuae authored May 10, 2024
```
Co-authored-by: miloice <jeffaw99@hotmail.com>
```
  ff5abcd7
02 May, 2024 1 commit
- [CI/Build] AMD CI pipeline with extended set of tests. (#4267) · 9b5c9f94
  Alexei-V-Ivanov-AMD authored May 02, 2024
```
Co-authored-by: simon-mo <simon.mo@hey.com>
```
  9b5c9f94
22 Apr, 2024 1 commit
- [AMD][Hardware][Misc][Bugfix] xformer cleanup and light navi logic and CI... · 95e5b087
  Hongxia Yang authored Apr 22, 2024
```
[AMD][Hardware][Misc][Bugfix] xformer cleanup and light navi logic and CI fixes and refactoring (#4129)
```
  95e5b087
10 Apr, 2024 1 commit
- [Bugfix][ROCm] Add numba to Dockerfile.rocm (#3962) · bd3c144e
  Woosuk Kwon authored Apr 10, 2024
  
  bd3c144e
09 Apr, 2024 1 commit

[ROCm][Hardware][AMD] Use Triton Kernel for default FA on ROCm (#3643) · 6c0b0451

Juan Villamizar authored Apr 09, 2024


Co-authored-by: jpvillam <jpvillam@amd.com>
Co-authored-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

6c0b0451

29 Mar, 2024 1 commit
- [ROCm][Bugfix] Fixed several bugs related to rccl path and attention selector logic (#3699) · 9765b5c4
  Hongxia Yang authored Mar 29, 2024
  
  9765b5c4
27 Mar, 2024 1 commit
- [Core] remove cupy dependency (#3625) · 8f44facd
  youkaichao authored Mar 27, 2024
  
  8f44facd
05 Mar, 2024 1 commit
- [ROCm] enable cupy in order to enable cudagraph mode for AMD GPUs (#3123) · 05af6da8
  Hongxia Yang authored Mar 04, 2024
```
Co-authored-by: lcskrishna <lollachaitanya@gmail.com>
```
  05af6da8
15 Feb, 2024 1 commit
- [ROCm] Dockerfile fix for flash-attention build (#2885) · 5255d99d
  Hongxia Yang authored Feb 15, 2024
  
  5255d99d
11 Feb, 2024 1 commit
- [ROCm] support Radeon™ 7900 series (gfx1100) without using flash-attention (#2768) · 0580aab0
  Hongxia Yang authored Feb 11, 2024
  
  0580aab0
07 Feb, 2024 1 commit
- [ROCm] Fix build problem resulted from previous commit related to FP8 kv-cache support (#2790) · c81dddb4
  Hongxia Yang authored Feb 07, 2024
  
  c81dddb4
05 Feb, 2024 1 commit
- [ROCm] Fixup arch checks for ROCM (#2627) · 2ccee3de
  Douglas Lehr authored Feb 05, 2024
  
  2ccee3de
26 Jan, 2024 1 commit
- [ROCm] add support to ROCm 6.0 and MI300 (#2274) · 6b7de1a0
  Hongxia Yang authored Jan 26, 2024
  
  6b7de1a0
14 Dec, 2023 1 commit
- Fix Dockerfile.rocm (#2101) · 05bdf4ea
  TJian authored Dec 14, 2023
```
Co-authored-by: miloice <jeffaw99@hotmail.com>
```
  05bdf4ea
13 Dec, 2023 1 commit
- [ROCm] Upgrade xformers version for ROCm & update doc (#2079) · f375ec84
  TJian authored Dec 13, 2023
```
Co-authored-by: miloice <jeffaw99@hotmail.com>
```
  f375ec84
08 Dec, 2023 1 commit

Merge EmbeddedLLM/vllm-rocm into vLLM main (#1836) · 6ccc0bff

TJian authored Dec 08, 2023


Co-authored-by: Philipp Moritz <pcmoritz@gmail.com>
Co-authored-by: Amir Balwel <amoooori04@gmail.com>
Co-authored-by: root <kuanfu.liu@akirakan.com>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: kuanfu <kuanfu.liu@embeddedllm.com>
Co-authored-by: miloice <17350011+kliuae@users.noreply.github.com>

6ccc0bff