docs: improve Fern sidebar titles, add guide subtitles, and replace ASCII diagrams with D2 (#6069)

Signed-off-by: Dan Gil <dagil@nvidia.com>
Signed-off-by: dagil-nvidia <dagil@nvidia.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Jonathan Tong <jt572@cornell.edu>

docs/diagrams/event_plane_coordination.d2

+direction: down
+
+coord: "Coordination Layer" {
+  style.font-size: 32
+  direction: right
+
+  sd: "Service Discovery" {
+    style.font-size: 28
+
+    k8s: "K8s: CRDs + API" {
+      style.font-size: 22
+      shape: rectangle
+    }
+    bm: "Bare metal: etcd" {
+      style.font-size: 22
+      shape: rectangle
+    }
+  }
+
+  nats: "NATS (Optional)" {
+    style.font-size: 28
+
+    kv: "KV Cache Events" {
+      style.font-size: 22
+      shape: rectangle
+    }
+    rs: "Router Replica Sync" {
+      style.font-size: 22
+      shape: rectangle
+    }
+    js: "JetStream Persistence" {
+      style.font-size: 22
+      shape: rectangle
+    }
+  }
+}
+
+frontend: Frontend {
+  style.font-size: 28
+  shape: rectangle
+}
+plan: Planner {
+  style.font-size: 28
+  shape: rectangle
+}
+worker: Worker {
+  style.font-size: 28
+  shape: rectangle
+}
+
+coord -> frontend
+coord -> plan
+coord -> worker

docs/diagrams/event_plane_lease.d2

+direction: right
+
+dr: "DistributedRuntime" {
+  style.font-size: 28
+
+  ns: "• Namespace" {
+    style.font-size: 22
+    shape: text
+  }
+  comp: "• Components" {
+    style.font-size: 22
+    shape: text
+  }
+  ep: "• Endpoints" {
+    style.font-size: 22
+    shape: text
+  }
+}
+
+lease: "Primary Lease\nTTL: 10s" {
+  style.font-size: 24
+  shape: rectangle
+  style.bold: true
+}
+
+etcd: etcd {
+  style.font-size: 28
+  shape: cylinder
+}
+
+dr -> lease
+lease -> etcd: "Keep-Alive\nHeartbeat" {
+  style.font-size: 22
+}

docs/diagrams/planner_architecture.d2

+direction: down
+
+planner: "Planner Component" {
+  style.font-size: 32
+
+  inputs: {
+    direction: right
+    style.border-radius: 8
+
+    mc: "Metric Collector\n(Prometheus)" {
+      style.font-size: 24
+      shape: rectangle
+    }
+    lp: "Load Predictor\n(ARIMA / Kalman / Prophet)" {
+      style.font-size: 24
+      shape: rectangle
+    }
+    pi: "Performance Interpolator\n(NPZ profiling data)" {
+      style.font-size: 24
+      shape: rectangle
+    }
+  }
+
+  sa: "Scaling Algorithm" {
+    style.font-size: 28
+    shape: rectangle
+    style.bold: true
+  }
+
+  connector: "Connector Layer" {
+    style.font-size: 28
+    direction: right
+
+    kc: "KubernetesConnector\n(PATCH DGD)" {
+      style.font-size: 24
+      shape: rectangle
+    }
+    vc: "VirtualConnector\n(Runtime bridge)" {
+      style.font-size: 24
+      shape: rectangle
+    }
+  }
+
+  inputs.mc -> sa
+  inputs.lp -> sa
+  inputs.pi -> sa
+  sa -> connector
+}

fern/pages/agents/tool-calling.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Connect Dynamo to external tools and services using function calling
 ---
 
 You can connect Dynamo to external tools and services using function calling (also known as tool calling). By providing a list of available functions, Dynamo can choose

fern/pages/benchmarks/benchmarking.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Benchmark and compare performance across Dynamo deployment configurations
 ---
 
 

fern/pages/components/kvbm/kvbm-guide.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Enable KV offloading using KV Block Manager (KVBM) for Dynamo deployments
 ---
 
 # KVBM Guide

fern/pages/components/router/router-guide.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Enable KV-aware routing using Router for Dynamo deployments
 ---
 
 # Router Guide

fern/pages/design-docs/event-plane.md

@@ -18,26 +18,7 @@ Dynamo's coordination layer adapts to the deployment environment:
 
 > **Note:** The runtime always defaults to `kv_store` (etcd) for service discovery. Kubernetes deployments must explicitly set `DYN_DISCOVERY_BACKEND=kubernetes` - the Dynamo operator handles this automatically.
 
-```
-┌─────────────────────────────────────────────────────────────────────┐
-│                    Coordination Layer                                │
-│                                                                      │
-│  ┌─────────────────────────┐    ┌─────────────────────────────────┐ │
-│  │   Service Discovery     │    │            NATS                 │ │
-│  │                         │    │         (Optional)              │ │
-│  │  • K8s: CRDs + API      │    │  • KV Cache Events              │ │
-│  │  • Bare metal: etcd     │    │  • Router Replica Sync          │ │
-│  │                         │    │  • JetStream Persistence        │ │
-│  └─────────────────────────┘    └─────────────────────────────────┘ │
-│                                                                      │
-└─────────────────────────────────────────────────────────────────────┘
-                    │                          │
-         ┌──────────┴──────────┐    ┌─────────┴──────────┐
-         ▼                     ▼    ▼                    ▼
-    ┌─────────┐          ┌─────────┐              ┌─────────┐
-    │Frontend │          │ Planner │              │ Worker  │
-    └─────────┘          └─────────┘              └─────────┘
-```
+![Coordination Layer showing Service Discovery and NATS connecting to Frontend, Planner, and Worker](/assets/img/event-plane-coordination.svg)
 
 ## Kubernetes-Native Service Discovery
 
@@ -102,19 +83,7 @@ export ETCD_ENDPOINTS=http://etcd-0:2379,http://etcd-1:2379,http://etcd-2:2379
 
 Each `DistributedRuntime` maintains a primary lease with etcd:
 
-```
-┌────────────────────┐         ┌──────────────┐
-│ DistributedRuntime │◄────────│ Primary Lease │
-│                    │         │  TTL: 10s     │
-│  • Namespace       │         └───────┬───────┘
-│  • Components      │                 │
-│  • Endpoints       │                 │ Keep-Alive
-│                    │                 │ Heartbeat
-└────────────────────┘                 ▼
-                               ┌──────────────┐
-                               │     etcd     │
-                               └──────────────┘
-```
+![DistributedRuntime connected to Primary Lease with Keep-Alive Heartbeat to etcd](/assets/img/event-plane-lease.svg)
 
 **Lease Lifecycle:**
 

fern/pages/design-docs/planner-design.md

@@ -13,30 +13,7 @@ The Planner is Dynamo's autoscaling controller. It observes system metrics, pred
 
 ## Architecture
 
-```text
-┌──────────────────────────────────────────────────────────┐
-│                    Planner Component                     │
-│                                                          │
-│  ┌───────────────┐ ┌───────────────┐ ┌────────────────┐  │
-│  │    Metric     │ │     Load      │ │  Performance   │  │
-│  │   Collector   │ │   Predictor   │ │  Interpolator  │  │
-│  │  (Prometheus) │ │ (ARIMA/etc.)  │ │  (JSON data)   │  │
-│  └───────┬───────┘ └───────┬───────┘ └───────┬────────┘  │
-│          │                 │                  │          │
-│          ▼                 ▼                  ▼          │
-│  ┌───────────────────────────────────────────────────┐   │
-│  │              Scaling Algorithm                    │   │
-│  └───────────────────────┬───────────────────────────┘   │
-│                          │                               │
-│  ┌───────────────────────▼───────────────────────────┐   │
-│  │               Connector Layer                     │   │
-│  │  ┌───────────────────┐  ┌───────────────────────┐ │   │
-│  │  │ KubernetesConn.   │  │   VirtualConn.        │ │   │
-│  │  │ (PATCH DGD)       │  │   (Runtime bridge)    │ │   │
-│  │  └───────────────────┘  └───────────────────────┘ │   │
-│  └───────────────────────────────────────────────────┘   │
-└──────────────────────────────────────────────────────────┘
-```
+![Planner architecture showing Metric Collector, Load Predictor, and Performance Interpolator feeding into the Scaling Algorithm and Connector Layer](../assets/img/planner-architecture.svg)
 
 ## Scaling Algorithm
 

fern/pages/development/backend-guide.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Create custom Python workers and engines for Dynamo
 ---
 
 # Writing Python Workers in Dynamo

fern/pages/fault-tolerance/README.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Handle failures gracefully with request migration, cancellation, and graceful shutdown
 ---
 
 # Fault Tolerance

fern/pages/features/disaggregated-serving/README.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Find optimal prefill/decode configuration for disaggregated serving deployments
 ---
 
 # Disaggregated Serving Guide

fern/pages/features/lora/README.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Serve fine-tuned LoRA adapters with dynamic loading and routing in Dynamo
 ---
 
 # LoRA Adapters

fern/pages/features/multimodal/README.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Deploy multimodal models with image, video, and audio support in Dynamo
 ---
 
 # Multimodal Inference in Dynamo

fern/pages/observability/README.md

 ---
 # SPDX-FileCopyrightText: Copyright (c) 2025-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
+subtitle: Monitor Dynamo deployments with metrics, logging, and tracing
 ---
 
 # Dynamo Observability

fern/versions/next.yml

@@ -78,11 +78,11 @@ navigation:
     contents:
       - page: KV Cache Aware Routing
         path: ../pages/components/router/router-guide.md
-      - page: Disaggregated Serving Guide
+      - page: Disaggregated Serving
         path: ../pages/features/disaggregated-serving/README.md
       - page: KV Cache Offloading
         path: ../pages/components/kvbm/kvbm-guide.md
-      - page: Dynamo Benchmarking Guide
+      - page: Dynamo Benchmarking
         path: ../pages/benchmarks/benchmarking.md
       - section: Multimodality Support
         path: ../pages/features/multimodal/README.md