refactor: move engine configs out of components directory (#3772)

Signed-off-by: Anant Sharma <anants@nvidia.com>
Co-authored-by: tanmayv25 <tanmay2592@gmail.com>

components/backends/trtllm/engine_configs/deepseek_r1/wide_ep/wide_ep_decode.yaml → recipes/deepseek-r1/trtllm/wide_ep/wide_ep_decode.yaml

@@ -17,7 +17,7 @@ backend: pytorch
 # WideEP related settings
 moe_config:
   backend: WIDEEP
-  load_balancer: /mnt/engine_configs/deepseek_r1/wide_ep/eplb.yaml
+  load_balancer: /mnt/recipes/deepseek-r1/trtllm/wide_ep/eplb.yaml
 
 # TP/EP/PP/DP
 tensor_parallel_size: 16

components/backends/trtllm/engine_configs/deepseek_r1/wide_ep/wide_ep_prefill.yaml → recipes/deepseek-r1/trtllm/wide_ep/wide_ep_prefill.yaml

@@ -17,7 +17,7 @@ backend: pytorch
 # WideEP related settings
 moe_config:
   backend: WIDEEP
-  load_balancer: /mnt/engine_configs/deepseek_r1/wide_ep/eplb.yaml
+  load_balancer: /mnt/recipes/deepseek-r1/trtllm/wide_ep/eplb.yaml
 
 # TP/EP/PP/DP
 tensor_parallel_size: 16

recipes/qwen2-vl-7b-instruct/trtllm/agg.yaml

+# 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.
+tensor_parallel_size: 8
+moe_expert_parallel_size: 1
+enable_attention_dp: false
+max_num_tokens: 4096
+max_batch_size: 8
+trust_remote_code: true
+backend: pytorch
+enable_chunked_prefill: true
+
+kv_cache_config:
+  free_gpu_memory_fraction: 0.3
+  enable_block_reuse: false
+
+cache_transceiver_config:
+  backend: DEFAULT
+# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
+# NOTE: overlap_scheduler enabled by default since this commit and changed
+# config field from 'enable_overlap_scheduler' to 'disable_overlap_scheduler':
+# https://github.com/NVIDIA/TensorRT-LLM/commit/b4e5df0ee0024eda3eeb83a6ba822245a30ab428

recipes/qwen3/trtllm/agg.yaml

+# 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.
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+enable_attention_dp: false
+max_num_tokens: 8192
+max_batch_size: 16
+trust_remote_code: true
+backend: pytorch
+enable_chunked_prefill: true
+
+kv_cache_config:
+  free_gpu_memory_fraction: 0.85
+
+# NOTE: pytorch_backend_config section flattened since: https://github.com/NVIDIA/TensorRT-LLM/pull/4603
+# NOTE: overlap_scheduler enabled by default since this commit and changed
+# config field from 'enable_overlap_scheduler' to 'disable_overlap_scheduler':
+# https://github.com/NVIDIA/TensorRT-LLM/commit/b4e5df0ee0024eda3eeb83a6ba822245a30ab428
+
+
+cuda_graph_config:
+  max_batch_size: 16
\ No newline at end of file

recipes/qwen3/trtllm/decode.yaml

+# 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.
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+enable_attention_dp: false
+max_num_tokens: 8192
+trust_remote_code: true
+backend: pytorch
+enable_chunked_prefill: true
+disable_overlap_scheduler: false
+
+cuda_graph_config:
+  max_batch_size: 16
+
+kv_cache_config:
+  free_gpu_memory_fraction: 0.85
+
+cache_transceiver_config:
+  backend: DEFAULT

recipes/qwen3/trtllm/prefill.yaml

+# 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.
+tensor_parallel_size: 1
+moe_expert_parallel_size: 1
+enable_attention_dp: false
+max_num_tokens: 8192
+trust_remote_code: true
+backend: pytorch
+enable_chunked_prefill: true
+# Overlap scheduler not currently supported in prefill only workers.
+disable_overlap_scheduler: true
+cuda_graph_config:
+  max_batch_size: 16
+kv_cache_config:
+  free_gpu_memory_fraction: 0.85
+
+cache_transceiver_config:
+  backend: DEFAULT
\ No newline at end of file