docs: Clean up incomplete recipes and clarify Kubernetes-only focus (#4159)

Signed-off-by: Ben Hamm <ben.hamm@gmail.com>
Signed-off-by: Tanmay Verma <tanmay2592@gmail.com>
Signed-off-by: atchernych <atchernych@nvidia.com>
Co-authored-by: Biswa Panda <biswa.panda@gmail.com>
Co-authored-by: tanmayv25 <tanmay2592@gmail.com>
Co-authored-by: Tanmay Verma <tanmayv@nvidia.com>
Co-authored-by: Anant Sharma <anants@nvidia.com>
Co-authored-by: atchernych <atchernych@nvidia.com>

examples/backends/trtllm/launch/agg.sh

@@ -6,7 +6,7 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen3-0.6B"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen3-0.6B"}
-export AGG_ENGINE_ARGS=${AGG_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/agg.yaml"}
+export AGG_ENGINE_ARGS=${AGG_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/agg.yaml"}
 export MODALITY=${MODALITY:-"text"}
 # If you want to use multimodal, set MODALITY to "multimodal"
 #export MODALITY=${MODALITY:-"multimodal"}

examples/backends/trtllm/launch/agg_metrics.sh

@@ -6,7 +6,7 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen3-0.6B"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen3-0.6B"}
-export AGG_ENGINE_ARGS=${AGG_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/agg.yaml"}
+export AGG_ENGINE_ARGS=${AGG_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/agg.yaml"}
 export MODALITY=${MODALITY:-"text"}
 
 # Setup cleanup trap

examples/backends/trtllm/launch/agg_router.sh

@@ -6,7 +6,7 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen3-0.6B"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen3-0.6B"}
-export AGG_ENGINE_ARGS=${AGG_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/agg.yaml"}
+export AGG_ENGINE_ARGS=${AGG_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/agg.yaml"}
 
 # Setup cleanup trap
 cleanup() {

examples/backends/trtllm/launch/disagg.sh

@@ -6,8 +6,8 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen3-0.6B"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen3-0.6B"}
-export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/prefill.yaml"}
-export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/decode.yaml"}
+export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/prefill.yaml"}
+export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/decode.yaml"}
 export PREFILL_CUDA_VISIBLE_DEVICES=${PREFILL_CUDA_VISIBLE_DEVICES:-"0"}
 export DECODE_CUDA_VISIBLE_DEVICES=${DECODE_CUDA_VISIBLE_DEVICES:-"1"}
 export MODALITY=${MODALITY:-"text"}

examples/backends/trtllm/launch/disagg_router.sh

@@ -6,8 +6,8 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen3-0.6B"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen3-0.6B"}
-export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/prefill.yaml"}
-export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen3/trtllm/decode.yaml"}
+export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/prefill.yaml"}
+export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen3/decode.yaml"}
 export PREFILL_CUDA_VISIBLE_DEVICES=${PREFILL_CUDA_VISIBLE_DEVICES:-"0"}
 export DECODE_CUDA_VISIBLE_DEVICES=${DECODE_CUDA_VISIBLE_DEVICES:-"1"}
 

examples/backends/trtllm/launch/disagg_same_gpu.sh

@@ -32,8 +32,8 @@ echo "GPU memory check passed: ${FREE_GPU_GB}GB available (required: ${REQUIRED_
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen3-0.6B"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen3-0.6B"}
-export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/tests/serve/configs/trtllm/prefill.yaml"}
-export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/tests/serve/configs/trtllm/decode.yaml"}
+export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/tests/serve/trtllm/engine_configs/qwen3/prefill.yaml"}
+export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/tests/serve/trtllm/engine_configs/qwen3/decode.yaml"}
 export CUDA_VISIBLE_DEVICES=${CUDA_VISIBLE_DEVICES:-"0"}
 export MODALITY=${MODALITY:-"text"}
 

examples/backends/trtllm/launch/epd_disagg.sh

@@ -6,9 +6,9 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"Qwen/Qwen2-VL-7B-Instruct"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"Qwen/Qwen2-VL-7B-Instruct"}
-export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen2-vl-7b-instruct/trtllm/prefill.yaml"}
-export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen2-vl-7b-instruct/trtllm/decode.yaml"}
-export ENCODE_ENGINE_ARGS=${ENCODE_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/qwen2-vl-7b-instruct/trtllm/encode.yaml"}
+export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen2-vl-7b-instruct/prefill.yaml"}
+export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen2-vl-7b-instruct/decode.yaml"}
+export ENCODE_ENGINE_ARGS=${ENCODE_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/qwen2-vl-7b-instruct/encode.yaml"}
 export PREFILL_CUDA_VISIBLE_DEVICES=${PREFILL_CUDA_VISIBLE_DEVICES:-"0"}
 export DECODE_CUDA_VISIBLE_DEVICES=${DECODE_CUDA_VISIBLE_DEVICES:-"1"}
 export ENCODE_CUDA_VISIBLE_DEVICES=${ENCODE_CUDA_VISIBLE_DEVICES:-"2"}

examples/backends/trtllm/launch/gpt_oss_disagg.sh

@@ -6,8 +6,8 @@
 export DYNAMO_HOME=${DYNAMO_HOME:-"/workspace"}
 export MODEL_PATH=${MODEL_PATH:-"/model"}
 export SERVED_MODEL_NAME=${SERVED_MODEL_NAME:-"openai/gpt-oss-120b"}
-export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/gpt-oss-120b/trtllm/disagg/prefill.yaml"}
-export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/recipes/gpt-oss-120b/trtllm/disagg/decode.yaml"}
+export PREFILL_ENGINE_ARGS=${PREFILL_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/gpt-oss-120b/prefill.yaml"}
+export DECODE_ENGINE_ARGS=${DECODE_ENGINE_ARGS:-"$DYNAMO_HOME/examples/backends/trtllm/engine_configs/gpt-oss-120b/decode.yaml"}
 
 set -e
 trap 'echo Cleaning up...; kill 0' EXIT

examples/basics/multinode/trtllm/README.md

+<!--
+SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+SPDX-License-Identifier: Apache-2.0
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+-->
+
+# Example: Multi-node TRTLLM Workers with Dynamo on Slurm
+
+See [here](/docs/backends/trtllm/multinode) for how to setup this example.

examples/basics/multinode/trtllm/srun_aggregated.sh

@@ -18,7 +18,7 @@ MOUNTS="${MOUNTS:-${DEFAULT_MOUNT}}"
 NUM_NODES=${NUM_NODES:-4}
 NUM_GPUS_PER_NODE=${NUM_GPUS_PER_NODE:-4}
 
-export ENGINE_CONFIG="${ENGINE_CONFIG:-/mnt/recipes/deepseek-r1/trtllm/agg/wide_ep/wide_ep_agg.yaml}"
+export ENGINE_CONFIG="${ENGINE_CONFIG:-/mnt/examples/backends/trtllm/engine_configs/deepseek-r1/agg/wide_ep/wide_ep_agg.yaml}"
 
 # Automate settings of certain variables for convenience, but you are free
 # to manually set these for more control as well.