Skip to content

Actions: ROCm/vllm

yapf

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
870 workflow runs
870 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Merge branch 'main' of github.com:ROCm/vllm
yapf #45: Commit d76bfa2 pushed by gshtras
March 26, 2024 17:44 49s main
March 26, 2024 17:44 49s
Hotfix: Disable BF16 opts due to ROCm 5.7 incompatibility
yapf #44: Commit f8f77d9 pushed by mawong-amd
March 26, 2024 16:57 44s main
March 26, 2024 16:57 44s
Re-apply the blockReduceSum fix for warp divergence
yapf #43: Commit c109d8a pushed by mawong-amd
March 26, 2024 14:59 47s main
March 26, 2024 14:59 47s
Refactor vector type to enable optimizations for bf16
yapf #42: Commit b036bac pushed by mawong-amd
March 26, 2024 14:20 54s main
March 26, 2024 14:20 54s
Update ROCm_performance.md
yapf #41: Commit 6c34e32 pushed by gshtras
March 25, 2024 21:14 46s main
March 25, 2024 21:14 46s
March 25, 2024 20:04 44s
Add a few missing unroll directives
yapf #39: Commit a7164ca pushed by mawong-amd
March 25, 2024 15:47 45s main
March 25, 2024 15:47 45s
March 25, 2024 14:58 48s
March 24, 2024 04:16 47s
Layernorm optimizations:
yapf #36: Commit 2d1baac pushed by mawong-amd
March 23, 2024 10:59 42s main
March 23, 2024 10:59 42s
Layernorm optimizations
yapf #35: Pull request #8 synchronize by mawong-amd
March 23, 2024 10:57 51s layernorm_fp16_opt
March 23, 2024 10:57 51s
Layernorm optimizations
yapf #34: Pull request #8 synchronize by mawong-amd
March 23, 2024 10:51 46s layernorm_fp16_opt
March 23, 2024 10:51 46s
Layernorm optimizations
yapf #33: Pull request #8 synchronize by mawong-amd
March 23, 2024 10:38 46s layernorm_fp16_opt
March 23, 2024 10:38 46s
Merge pull request #7 from ROCm/integration_no_fp8
yapf #32: Commit 629f74b pushed by gshtras
March 22, 2024 22:25 46s main
March 22, 2024 22:25 46s
Features integration without fp8
yapf #31: Pull request #7 synchronize by gshtras
March 22, 2024 20:59 46s integration_no_fp8
March 22, 2024 20:59 46s
Features integration without fp8
yapf #30: Pull request #7 synchronize by jpvillam-amd
March 22, 2024 20:11 45s integration_no_fp8
March 22, 2024 20:11 45s
Adding new rocm triton flash attention kernel
yapf #29: Pull request #4 synchronize by jpvillam-amd
March 22, 2024 20:00 51s jpvillam/v0.3.3_triton
March 22, 2024 20:00 51s
Adding new rocm triton flash attention kernel
yapf #28: Pull request #4 synchronize by jpvillam-amd
March 22, 2024 19:08 46s jpvillam/v0.3.3_triton
March 22, 2024 19:08 46s
Features integration without fp8
yapf #27: Pull request #7 synchronize by gshtras
March 22, 2024 18:06 54s integration_no_fp8
March 22, 2024 18:06 54s
Features integration without fp8
yapf #26: Pull request #7 synchronize by gshtras
March 22, 2024 17:43 1m 9s integration_no_fp8
March 22, 2024 17:43 1m 9s
Features integration without fp8
yapf #25: Pull request #7 synchronize by gshtras
March 22, 2024 16:51 44s integration_no_fp8
March 22, 2024 16:51 44s
Adding new rocm triton flash attention kernel
yapf #24: Pull request #4 synchronize by jpvillam-amd
March 22, 2024 16:49 44s jpvillam/v0.3.3_triton
March 22, 2024 16:49 44s
Features integration
yapf #23: Pull request #5 synchronize by gshtras
March 22, 2024 00:06 50s integration
March 22, 2024 00:06 50s
Features integration without fp8
yapf #22: Pull request #7 opened by gshtras
March 21, 2024 23:58 47s integration_no_fp8
March 21, 2024 23:58 47s
Features integration
yapf #21: Pull request #5 synchronize by gshtras
March 21, 2024 21:16 49s integration
March 21, 2024 21:16 49s
ProTip! You can narrow down the results and go further in time using created:<2024-03-21 or the other filters available.