Skip to content

Commit

Permalink
Add HPU support to benchmark_latency and benchmark_throughput (#49)
Browse files Browse the repository at this point in the history
  • Loading branch information
kzawora-intel authored and adobrzyniewicz-habana committed Jun 25, 2024
1 parent 6948b7d commit e74354a
Show file tree
Hide file tree
Showing 2 changed files with 4 additions and 4 deletions.
4 changes: 2 additions & 2 deletions benchmarks/benchmark_latency.py
Original file line number Diff line number Diff line change
Expand Up @@ -170,8 +170,8 @@ def run_to_completion(profile_dir: Optional[str] = None):
"--device",
type=str,
default="cuda",
choices=["cuda", "cpu"],
help='device type for vLLM execution, supporting CUDA and CPU.')
choices=["cuda", "cpu", "hpu"],
help='device type for vLLM execution, supporting CUDA, CPU and HPU.')
parser.add_argument('--block-size',
type=int,
default=16,
Expand Down
4 changes: 2 additions & 2 deletions benchmarks/benchmark_throughput.py
Original file line number Diff line number Diff line change
Expand Up @@ -334,8 +334,8 @@ def main(args: argparse.Namespace):
"--device",
type=str,
default="cuda",
choices=["cuda", "cpu"],
help='device type for vLLM execution, supporting CUDA and CPU.')
choices=["cuda", "cpu", "hpu"],
help='device type for vLLM execution, supporting CUDA, CPU and HPU.')
parser.add_argument(
"--enable-prefix-caching",
action='store_true',
Expand Down

0 comments on commit e74354a

Please sign in to comment.