docs: cleanup of docs refactor for components, integrations, and features

README.md

@@ -52,10 +52,10 @@ Built in Rust for performance and Python for extensibility, Dynamo is fully open
 |---|:----:|:----------:|:--:|
 | **Best For** | High-throughput serving | Maximum performance | Broadest feature coverage |
 | [**Disaggregated Serving**](docs/design_docs/disagg_serving.md) | ✅ | ✅ | ✅ |
-| [**KV-Aware Routing**](docs/router/README.md) | ✅ | ✅ | ✅ |
-| [**SLA-Based Planner**](docs/planner/sla_planner.md) | ✅ | ✅ | ✅ |
-| [**KVBM**](docs/kvbm/README.md) | 🚧 | ✅ | ✅ |
-| [**Multimodal**](docs/multimodal/index.md) | ✅ | ✅ | ✅ |
+| [**KV-Aware Routing**](docs/components/router/README.md) | ✅ | ✅ | ✅ |
+| [**SLA-Based Planner**](docs/components/planner/planner_guide.md) | ✅ | ✅ | ✅ |
+| [**KVBM**](docs/components/kvbm/README.md) | 🚧 | ✅ | ✅ |
+| [**Multimodal**](docs/features/multimodal/README.md) | ✅ | ✅ | ✅ |
 | [**Tool Calling**](docs/agents/tool-calling.md) | ✅ | ✅ | ✅ |
 
 > **[Full Feature Matrix →](docs/reference/feature-matrix.md)** — Detailed compatibility including LoRA, Request Migration, Speculative Decoding, and feature interactions.
@@ -347,7 +347,7 @@ python3 -m dynamo.frontend
 Dynamo provides comprehensive benchmarking tools:
 
 - **[Benchmarking Guide](docs/benchmarks/benchmarking.md)** – Compare deployment topologies using AIPerf
-- **[SLA-Driven Deployments](docs/planner/sla_planner_quickstart.md)** – Optimize deployments to meet SLA requirements
+- **[SLA-Driven Deployments](docs/components/planner/planner_guide.md)** – Optimize deployments to meet SLA requirements
 
 ## Frontend OpenAPI Specification
 
@@ -357,7 +357,7 @@ The OpenAI-compatible frontend exposes an OpenAPI 3 spec at `/openapi.json`. To
 cargo run -p dynamo-llm --bin generate-frontend-openapi
 ```
 
-This writes to `docs/frontends/openapi.json`.
+This writes to `docs/reference/api/openapi.json`.
 
 ## Service Discovery and Messaging
 
@@ -388,9 +388,9 @@ See [SGLang on Slurm](examples/backends/sglang/slurm_jobs/README.md) and [TRT-LL
 
 <!-- Reference links for Feature Compatibility Matrix -->
 [disagg]: docs/design_docs/disagg_serving.md
-[kv-routing]: docs/router/README.md
-[planner]: docs/planner/sla_planner.md
-[kvbm]: docs/kvbm/README.md
+[kv-routing]: docs/components/router/README.md
+[planner]: docs/components/planner/planner_guide.md
+[kvbm]: docs/components/kvbm/README.md
 [mm]: examples/multimodal/
 [migration]: docs/fault_tolerance/request_migration.md
 [lora]: examples/backends/vllm/deploy/lora/README.md

benchmarks/profiler/README.md

@@ -0,0 +1,13 @@
+<!--
+SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES.
+All rights reserved.
+SPDX-License-Identifier: Apache-2.0
+-->
+
+# Profiler
+
+Documentation for the Dynamo Profiler has moved to [docs/components/profiler/](../../docs/components/profiler/README.md).
+
+- [Profiler Overview](../../docs/components/profiler/README.md)
+- [Profiler Guide](../../docs/components/profiler/profiler_guide.md)
+- [Profiler Examples](../../docs/components/profiler/profiler_examples.md)

benchmarks/profiler/webui/utils.py

@@ -620,7 +620,7 @@ def create_gradio_interface(
 
             > 📝 **Note:** The dotted red line in the prefill and decode charts are default TTFT and ITL SLAs if not specified.
 
-            > ⚠️ **Warning:** The TTFT values here represent the ideal case when requests arrive uniformly, minimizing queueing. Real-world TTFT may be higher than profiling results. To mitigate the issue, planner uses [correction factors](https://github.com/ai-dynamo/dynamo/blob/main/docs/planner/sla_planner.md#2-correction-factor-calculation) to adjust dynamically at runtime.
+            > ⚠️ **Warning:** The TTFT values here represent the ideal case when requests arrive uniformly, minimizing queueing. Real-world TTFT may be higher than profiling results. To mitigate the issue, planner uses [correction factors](https://github.com/ai-dynamo/dynamo/blob/main/docs/design_docs/planner_design.md#step-2-correction-factor-calculation) to adjust dynamically at runtime.
 
             > 💡 **Tip:** Use the GPU cost checkbox and input in the charts section to convert GPU hours to cost.
             """

benchmarks/router/README.md

@@ -127,7 +127,7 @@ To see all available router arguments, run:
 python -m dynamo.frontend --help
 ```
 
-For detailed explanations of router arguments (especially KV cache routing parameters), see the [Router Guide](../../docs/router/router_guide.md).
+For detailed explanations of router arguments (especially KV cache routing parameters), see the [Router Guide](../../docs/components/router/router_guide.md).
 
 > [!Note]
 > If you're unsure whether your backend engines correctly emit KV events for certain models (e.g., hybrid models like gpt-oss or nemotron nano 2), use the `--no-kv-events` flag to disable KV event tracking and use approximate KV indexing instead:
@@ -146,7 +146,7 @@ When you launch prefill workers using `run_engines.sh --prefill`, the frontend a
 - Uses the same routing mode as the frontend's `--router-mode` setting
 - Seamlessly integrates with your decode workers for token generation
 
-No additional configuration is needed - simply launch both decode and prefill workers, and the system handles the rest. See the [Router Guide](../../docs/router/router_guide.md#disaggregated-serving) for more details.
+No additional configuration is needed - simply launch both decode and prefill workers, and the system handles the rest. See the [Router Guide](../../docs/components/router/router_guide.md#disaggregated-serving) for more details.
 
 > [!Note]
 > The unified frontend with automatic prefill routing is currently enabled for vLLM and TensorRT-LLM backends. For SGLang (work in progress), you need to launch a separate standalone router as the prefill router targeting the prefill endpoints. See example script: [`examples/backends/sglang/launch/disagg_router.sh`](../../examples/backends/sglang/launch/disagg_router.sh)

components/src/dynamo/mocker/README.md

@@ -60,7 +60,7 @@ python -m dynamo.mocker \
 
 The profile results directory should contain `selected_prefill_interpolation/` and `selected_decode_interpolation/` subdirectories with `raw_data.npz` files. This works seamlessly in Kubernetes where profile data is mounted via ConfigMap or PersistentVolume.
 
-To generate profiling data for your own model/hardware configuration, run the profiler (see [SLA-driven profiling documentation](../../../../docs/benchmarks/sla_driven_profiling.md) for details):
+To generate profiling data for your own model/hardware configuration, run the profiler (see [SLA-driven profiling documentation](../../../../docs/components/profiler/profiler_guide.md) for details):
 
 ```bash
 python benchmarks/profiler/profile_sla.py \

components/src/dynamo/planner/README.md

@@ -19,5 +19,5 @@ limitations under the License.
 
 SLA-driven autoscaling controller for Dynamo inference graphs.
 
-- **User docs**: [docs/planner/](/docs/planner/) (deployment, configuration, examples)
+- **User docs**: [docs/planner/](/docs/components/planner/) (deployment, configuration, examples)
 - **Design docs**: [docs/design_docs/planner_design.md](/docs/design_docs/planner_design.md) (architecture, algorithms)

components/src/dynamo/planner/utils/perf_interpolation.py

@@ -29,7 +29,7 @@
 
 MISSING_PROFILING_DATA_ERROR_MESSAGE = (
     "SLA-Planner requires pre-deployment profiling results to run.\n"
-    "Please follow /docs/benchmarks/sla_driven_profiling.md to run the profiling first,\n"
+    "Please follow /docs/components/profiler/profiler_guide.md to run the profiling first,\n"
     "and make sure the profiling results are present in --profile-results-dir."
 )
 

components/src/dynamo/router/README.md

@@ -3,7 +3,7 @@
 
 # Standalone Router
 
-A backend-agnostic standalone KV-aware router service for Dynamo deployments. For details on how KV-aware routing works, see the [Router Guide](/docs/router/router_guide.md).
+A backend-agnostic standalone KV-aware router service for Dynamo deployments. For details on how KV-aware routing works, see the [Router Guide](/docs/components/router/router_guide.md).
 
 ## Overview
 
@@ -29,7 +29,7 @@ python -m dynamo.router \
 - `--endpoint`: Full endpoint path for workers in the format `namespace.component.endpoint` (e.g., `dynamo.prefill.generate`)
 
 **Router Configuration:**
-For detailed descriptions of all KV router configuration options including `--block-size`, `--kv-overlap-score-weight`, `--router-temperature`, `--no-kv-events`, `--router-replica-sync`, `--router-snapshot-threshold`, `--router-reset-states`, and `--no-track-active-blocks`, see the [Router Guide](/docs/router/router_guide.md).
+For detailed descriptions of all KV router configuration options including `--block-size`, `--kv-overlap-score-weight`, `--router-temperature`, `--no-kv-events`, `--router-replica-sync`, `--router-snapshot-threshold`, `--router-reset-states`, and `--no-track-active-blocks`, see the [Router Guide](/docs/components/router/router_guide.md).
 
 ## Architecture
 
@@ -43,7 +43,7 @@ Clients query the `find_best_worker` endpoint to determine which worker should p
 ## Example: Manual Disaggregated Serving (Alternative Setup)
 
 > [!Note]
-> **This is an alternative advanced setup.** The recommended approach for disaggregated serving is to use the frontend's automatic prefill routing, which activates when you register workers with `ModelType.Prefill`. See the [Router Guide](/docs/router/router_guide.md#disaggregated-serving) for the default setup.
+> **This is an alternative advanced setup.** The recommended approach for disaggregated serving is to use the frontend's automatic prefill routing, which activates when you register workers with `ModelType.Prefill`. See the [Router Guide](/docs/components/router/router_guide.md#disaggregated-serving) for the default setup.
 >
 > Use this manual setup if you need explicit control over prefill routing configuration or want to manage prefill and decode routers separately.
 
@@ -103,7 +103,7 @@ See [`components/src/dynamo/vllm/handlers.py`](../vllm/handlers.py) for a refere
 
 ## See Also
 
-- [Router Guide](/docs/router/router_guide.md) - Configuration and tuning for KV-aware routing
+- [Router Guide](/docs/components/router/router_guide.md) - Configuration and tuning for KV-aware routing
 - [Router Design](/docs/design_docs/router_design.md) - Architecture details and event transport modes
 - [Frontend Router](../frontend/README.md) - Main HTTP frontend with integrated routing
 - [Router Benchmarking](/benchmarks/router/README.md) - Performance testing and tuning

deploy/inference-gateway/README.md

@@ -220,7 +220,7 @@ Common Vars for Routing Configuration:
   - Set `DYNAMO_OVERLAP_SCORE_WEIGHT` to weigh how heavily the score uses token overlap (predicted KV cache hits) versus other factors (load, historical hit rate). Higher weight biases toward reusing workers with similar cached prefixes.
   - Set `DYNAMO_ROUTER_TEMPERATURE` to soften or sharpen the selection curve when combining scores. Low temperature makes the router pick the top candidate deterministically; higher temperature lets lower-scoring workers through more often (exploration).
   - Set `DYNAMO_USE_KV_EVENTS=false` if you want to disable the workers sending KV events while using kv-routing
-  - See the [Router Guide](../../docs/router/router_guide.md) for details.
+  - See the [Router Guide](../../docs/components/router/router_guide.md) for details.
 
 
 Stand-Alone installation only:

deploy/utils/README.md

@@ -145,7 +145,7 @@ kubectl delete pod pvc-access-pod -n $NAMESPACE
 
 For complete benchmarking and profiling workflows:
 - **Benchmarking Guide**: See [docs/benchmarks/benchmarking.md](../../docs/benchmarks/benchmarking.md) for comparing DynamoGraphDeployments and external endpoints
-- **Pre-Deployment Profiling**: See [docs/benchmarks/sla_driven_profiling.md](../../docs/benchmarks/sla_driven_profiling.md) for optimizing configurations before deployment
+- **Pre-Deployment Profiling**: See [docs/components/profiler/profiler_guide.md](../../docs/components/profiler/profiler_guide.md) for optimizing configurations before deployment
 
 ## Notes
 

docs/api/nixl_connect/README.md

@@ -103,7 +103,7 @@ flowchart LR
 
 ### Multimodal Example
 
-In the case of the [Dynamo Multimodal Disaggregated Example](../../multimodal/vllm.md):
+In the case of the [Dynamo Multimodal Disaggregated Example](../../features/multimodal/multimodal_vllm.md):
 
  1. The HTTP frontend accepts a text prompt and a URL to an image.
 

docs/backends/sglang/README.md

@@ -36,10 +36,10 @@ git checkout $(git describe --tags $(git rev-list --tags --max-count=1))
 |---------|--------|-------|
 | [**Disaggregated Serving**](../../design_docs/disagg_serving.md) | ✅ |  |
 | [**Conditional Disaggregation**](../../design_docs/disagg_serving.md#conditional-disaggregation) | 🚧 | WIP [PR](https://github.com/sgl-project/sglang/pull/7730) |
-| [**KV-Aware Routing**](../../router/README.md) | ✅ |  |
-| [**SLA-Based Planner**](../../planner/sla_planner.md) | ✅ |  |
-| [**Multimodal Support**](../../multimodal/sglang.md) | ✅ |  |
-| [**KVBM**](../../kvbm/README.md) | ❌ | Planned |
+| [**KV-Aware Routing**](../../components/router/README.md) | ✅ |  |
+| [**SLA-Based Planner**](../../components/planner/planner_guide.md) | ✅ |  |
+| [**Multimodal Support**](../../features/multimodal/multimodal_sglang.md) | ✅ |  |
+| [**KVBM**](../../components/kvbm/README.md) | ❌ | Planned |
 
 
 ## Dynamo SGLang Integration

docs/backends/trtllm/README.md

@@ -55,10 +55,10 @@ git checkout $(git describe --tags $(git rev-list --tags --max-count=1))
 |---------|--------------|-------|
 | [**Disaggregated Serving**](../../../docs/design_docs/disagg_serving.md) | ✅ |  |
 | [**Conditional Disaggregation**](../../../docs/design_docs/disagg_serving.md#conditional-disaggregation) | 🚧 | Not supported yet |
-| [**KV-Aware Routing**](../../router/README.md) | ✅ |  |
-| [**SLA-Based Planner**](../../../docs/planner/sla_planner.md) | ✅ |  |
-| [**Load Based Planner**](../../../docs/planner/load_planner.md) | 🚧 | Planned |
-| [**KVBM**](../../../docs/kvbm/README.md) | ✅ | |
+| [**KV-Aware Routing**](../../components/router/README.md) | ✅ |  |
+| [**SLA-Based Planner**](../../../docs/components/planner/planner_guide.md) | ✅ |  |
+| [**Load Based Planner**](../../../docs/components/planner/README.md) | 🚧 | Planned |
+| [**KVBM**](../../../docs/components/kvbm/README.md) | ✅ | |
 
 ### Large Scale P/D and WideEP Features
 
@@ -114,7 +114,7 @@ apt-get update && apt-get -y install git git-lfs
 > [!IMPORTANT]
 > Below we provide some simple shell scripts that run the components for each configuration. Each shell script is simply running the `python3 -m dynamo.frontend <args>` to start up the ingress and using `python3 -m dynamo.trtllm <args>` to start up the workers. You can easily take each command and run them in separate terminals.
 
-For detailed information about the architecture and how KV-aware routing works, see the [Router Guide](../../router/router_guide.md).
+For detailed information about the architecture and how KV-aware routing works, see the [Router Guide](../../components/router/router_guide.md).
 
 ### Aggregated
 ```bash
@@ -231,7 +231,7 @@ To benchmark your deployment with AIPerf, see this utility script, configuring t
 
 ## Multimodal support
 
-Dynamo with the TensorRT-LLM backend supports multimodal models, enabling you to process both text and images (or pre-computed embeddings) in a single request. For detailed setup instructions, example requests, and best practices, see the [TensorRT-LLM Multimodal Guide](../../multimodal/trtllm.md).
+Dynamo with the TensorRT-LLM backend supports multimodal models, enabling you to process both text and images (or pre-computed embeddings) in a single request. For detailed setup instructions, example requests, and best practices, see the [TensorRT-LLM Multimodal Guide](../../features/multimodal/multimodal_trtllm.md).
 
 ## Logits Processing
 
@@ -327,7 +327,7 @@ For detailed instructions on running comprehensive performance sweeps across bot
 
 Dynamo with TensorRT-LLM currently supports integration with the Dynamo KV Block Manager. This integration can significantly reduce time-to-first-token (TTFT) latency, particularly in usage patterns such as multi-turn conversations and repeated long-context requests.
 
-Here is the instruction: [Running KVBM in TensorRT-LLM](./../../../docs/kvbm/kvbm_guide.md#run-kvbm-in-dynamo-with-tensorrt-llm) .
+Here is the instruction: [Running KVBM in TensorRT-LLM](./../../../docs/components/kvbm/kvbm_guide.md#run-kvbm-in-dynamo-with-tensorrt-llm) .
 
 ## Known Issues and Mitigations