Initialize collective profiler with model config

runs/frontier-qwen30-vllm020-profile-v1/profile_vllm020_allreduce.py

@@ -23,6 +23,7 @@ VLLM_COMMIT = "88d34c6409e9fb3c7b8ca0c04756f061d2099eb1"
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser()
     parser.add_argument("--vllm-source", type=Path, required=True)
+    parser.add_argument("--model", type=Path, required=True)
     parser.add_argument("--output", type=Path, required=True)
     parser.add_argument("--num-tokens", type=int, nargs="+", default=[8])
     parser.add_argument("--hidden-dim", type=int, default=2048)
@@ -69,7 +70,12 @@ def main() -> None:
         tensor_model_parallel_all_reduce,
     )
     from vllm.distributed.parallel_state import get_tp_group
-    from vllm.config import ParallelConfig, VllmConfig, set_current_vllm_config
+    from vllm.config import (
+        ModelConfig,
+        ParallelConfig,
+        VllmConfig,
+        set_current_vllm_config,
+    )
 
     rank = int(os.environ["RANK"])
     local_rank = int(os.environ["LOCAL_RANK"])
@@ -80,7 +86,15 @@ def main() -> None:
     torch.accelerator.set_device_index(device)
     torch.set_default_device(device)
     init_distributed_environment()
+    model_config = ModelConfig(
+        model=str(args.model),
+        dtype="bfloat16",
+        max_model_len=8192,
+        skip_tokenizer_init=True,
+        generation_config="vllm",
+    )
     vllm_config = VllmConfig(
+        model_config=model_config,
         parallel_config=ParallelConfig(tensor_parallel_size=world_size)
     )
     with set_current_vllm_config(vllm_config):
@@ -162,6 +176,7 @@ def main() -> None:
                 "torch_version": torch.__version__,
                 "torch_cuda": torch.version.cuda,
                 "gpu": torch.cuda.get_device_name(device),
+                "model": str(args.model),
                 "backend_env": {
                     "VLLM_ALLREDUCE_USE_FLASHINFER": "1",
                     "VLLM_FLASHINFER_ALLREDUCE_BACKEND": "trtllm",