enable GPUDirect support for intel xpu on Dynamo

Signed-off-by: Spycsh <sihan.chen@intel.com>

Author: Spycsh

components/src/dynamo/vllm/multimodal_utils/chat_processor.py

@@ -22,7 +22,11 @@
 from vllm.entrypoints.chat_utils import ConversationMessage
 from vllm.inputs.data import TokensPrompt
 from vllm.sampling_params import SamplingParams
-from vllm.tokenizers import TokenizerLike as AnyTokenizer
+
+try:
+    from vllm.tokenizers import TokenizerLike as AnyTokenizer
+except ImportError:
+    from vllm.transformers_utils.tokenizer import AnyTokenizer
 
 # Try importing from new vLLM (https://github.com/vllm-project/vllm/pull/32369), fallback to old structure
 try:

examples/backends/vllm/launch/disagg_router_xpu_gdr.sh

@@ -0,0 +1,66 @@
+
+#!/bin/bash
+# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: Apache-2.0
+set -e
+trap 'echo Cleaning up...; kill 0' EXIT
+
+# Set deterministic hash for KV event IDs
+export PYTHONHASHSEED=0
+
+# Common configuration
+MODEL="Qwen/Qwen3-0.6B"
+BLOCK_SIZE=64
+VLLM_NIXL_DEVICE_TO_DEVICE=true
+NIXL_BUFFER_DEVICE=xpu
+VLLM_NIXL_BACKEND=UCX
+export UCX_MEMTYPE_CACHE=0
+export UCX_NET_DEVICES=mlx5_0:1,mlx5_1:1,mlx5_2:1,mlx5_3:1
+export UCX_TLS=ib,rc,ze_copy
+
+
+# Start frontend with KV routing
+# The frontend will automatically detect prefill workers and activate an internal prefill router
+# edit --router-mode to random / round-robin / kv
+python -m dynamo.frontend \
+    --router-mode kv \
+    --http-port 8000 \
+    --router-reset-states &
+
+# two decode workers
+VLLM_NIXL_SIDE_CHANNEL_PORT=20096 \
+ZE_AFFINITY_MASK=0 python3 -m dynamo.vllm \
+    --model $MODEL \
+    --block-size $BLOCK_SIZE \
+    --kv-transfer-config "{\"kv_connector\": \"NixlConnector\", \"kv_role\": \"kv_both\", \"kv_buffer_device\": \"${NIXL_BUFFER_DEVICE}\", \"kv_connector_extra_config\": {\"backends\": [\"${VLLM_NIXL_BACKEND}\"]}}" \
+    --connector none \
+    --kv-events-config '{"publisher":"zmq","topic":"kv-events","endpoint":"tcp://*:5556", "enable_kv_cache_events":true}' &
+
+VLLM_NIXL_SIDE_CHANNEL_PORT=20097 \
+ZE_AFFINITY_MASK=1 python3 -m dynamo.vllm \
+    --model $MODEL \
+    --block-size $BLOCK_SIZE \
+    --kv-transfer-config "{\"kv_connector\": \"NixlConnector\", \"kv_role\": \"kv_both\", \"kv_buffer_device\": \"${NIXL_BUFFER_DEVICE}\", \"kv_connector_extra_config\": {\"backends\": [\"${VLLM_NIXL_BACKEND}\"]}}" \
+    --connector none \
+    --kv-events-config '{"publisher":"zmq","topic":"kv-events","endpoint":"tcp://*:5557", "enable_kv_cache_events":true}' &
+
+# two prefill workers
+# When registered with --is-prefill-worker, these workers are automatically detected
+# by the frontend, which activates an internal prefill router for KV-aware prefill routing
+VLLM_NIXL_SIDE_CHANNEL_PORT=20098 \
+ZE_AFFINITY_MASK=2 python3 -m dynamo.vllm \
+    --model $MODEL \
+    --block-size $BLOCK_SIZE \
+    --kv-transfer-config "{\"kv_connector\": \"NixlConnector\", \"kv_role\": \"kv_both\", \"kv_buffer_device\": \"${NIXL_BUFFER_DEVICE}\", \"kv_connector_extra_config\": {\"backends\": [\"${VLLM_NIXL_BACKEND}\"]}}" \
+    --connector none \
+    --is-prefill-worker \
+    --kv-events-config '{"publisher":"zmq","topic":"kv-events","endpoint":"tcp://*:5558", "enable_kv_cache_events":true}' &
+
+VLLM_NIXL_SIDE_CHANNEL_PORT=20099 \
+ZE_AFFINITY_MASK=3 python3 -m dynamo.vllm \
+    --model $MODEL \
+    --block-size $BLOCK_SIZE \
+    --kv-transfer-config "{\"kv_connector\": \"NixlConnector\", \"kv_role\": \"kv_both\", \"kv_buffer_device\": \"${NIXL_BUFFER_DEVICE}\", \"kv_connector_extra_config\": {\"backends\": [\"${VLLM_NIXL_BACKEND}\"]}}" \
+    --connector none \
+    --is-prefill-worker \
+    --kv-events-config '{"publisher":"zmq","topic":"kv-events","endpoint":"tcp://*:5559", "enable_kv_cache_events":true}'