Skip to content

Commit 4112941

Browse files
authored
Merge branch 'main' into 1012-yuluo/chore-docs
2 parents fd971a4 + 4328db1 commit 4112941

File tree

5 files changed

+35
-24
lines changed

5 files changed

+35
-24
lines changed

website/docs/intro.md

Lines changed: 6 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -46,8 +46,10 @@ Our testing shows significant improvements in model accuracy through specialized
4646

4747
## 🛠️ Architecture Overview
4848

49-
```mermaid
50-
graph TB
49+
import ZoomableMermaid from '@site/src/components/ZoomableMermaid';
50+
51+
<ZoomableMermaid title="Architecture Overview" defaultZoom={3.1}>
52+
{`graph TB
5153
Client[Client Request] --> Envoy[Envoy Proxy]
5254
Envoy --> Router[Semantic Router ExtProc]
5355

@@ -74,8 +76,8 @@ graph TB
7476
Models --> Math[Math Model]
7577
Models --> Creative[Creative Model]
7678
Models --> Code[Code Model]
77-
Models --> General[General Model]
78-
```
79+
Models --> General[General Model]`}
80+
</ZoomableMermaid>
7981

8082
## 🎯 Use Cases
8183

website/docs/overview/mixture-of-models.md

Lines changed: 10 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -100,8 +100,10 @@ Different models excel at different tasks. MoM leverages this specialization:
100100

101101
#### 3. **Improved System Reliability**
102102

103-
```mermaid
104-
graph TB
103+
import ZoomableMermaid from '@site/src/components/ZoomableMermaid';
104+
105+
<ZoomableMermaid title="System Architecture Overview" defaultZoom={5.5}>
106+
{`graph TB
105107
subgraph "Single Model Risk"
106108
SingleQuery[Query] --> SingleModel[GPT-4]
107109
SingleModel -->|Failure| SingleFailure[Complete System Down]
@@ -114,8 +116,8 @@ graph TB
114116
Router --> Model3[Model C]
115117
Model1 -->|Failure| Fallback[Automatic Fallback]
116118
Fallback --> Model2
117-
end
118-
```
119+
end`}
120+
</ZoomableMermaid>
119121

120122
**Reliability Benefits:**
121123

@@ -276,8 +278,8 @@ subject_routing = {
276278

277279
MoM architecture supports various deployment strategies:
278280

279-
```mermaid
280-
graph TB
281+
<ZoomableMermaid title="System Architecture Overview" defaultZoom={5.5}>
282+
{`graph TB
281283
subgraph "Cloud Deployment"
282284
CloudQueries[Queries] --> CloudRouter[Cloud Router]
283285
CloudRouter --> OpenAI[OpenAI GPT]
@@ -295,8 +297,8 @@ graph TB
295297
OnPremQueries[Queries] --> OnPremRouter[On-Prem Router]
296298
OnPremRouter --> LocalLLaMA[Local LLaMA Models]
297299
OnPremRouter --> FineTuned[Fine-tuned Specialized Models]
298-
end
299-
```
300+
end`}
301+
</ZoomableMermaid>
300302

301303
### 2. **A/B Testing and Gradual Rollouts**
302304

website/docs/overview/semantic-router-overview.md

Lines changed: 6 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -151,8 +151,10 @@ GPT-5 introduces a revolutionary **router-as-coordinator** architecture:
151151

152152
**Operational Flow:**
153153

154-
```mermaid
155-
sequenceDiagram
154+
import ZoomableMermaid from '@site/src/components/ZoomableMermaid';
155+
156+
<ZoomableMermaid title="Operational Flow" defaultZoom={5.5}>
157+
{`sequenceDiagram
156158
participant User
157159
participant Router as GPT-5 Router
158160
participant Math as Math Specialist
@@ -164,8 +166,8 @@ sequenceDiagram
164166
Router->>Router: Analyze query intent
165167
Router->>Math: Route to math specialist
166168
Math->>Router: Mathematical solution
167-
Router->>User: Optimized response
168-
```
169+
Router->>User: Optimized response`}
170+
</ZoomableMermaid>
169171

170172
**Business Impact:**
171173

website/docs/proposals/nvidia-dynamo-integration.md

Lines changed: 7 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -528,8 +528,11 @@ prompt_guard:
528528

529529
### 4.3 System Architecture
530530

531-
```mermaid
532-
graph TB
531+
import ZoomableMermaid from '@site/src/components/ZoomableMermaid';
532+
533+
<ZoomableMermaid title="System Architecture Overview" defaultZoom={10.5}>
534+
535+
{`graph TB
533536
Client[LLM Application<br/>OpenAI SDK]
534537

535538
subgraph Main["Main Processing Flow"]
@@ -628,8 +631,8 @@ graph TB
628631
style DynamoRouter fill:#c8e6c9
629632
style SemanticCache fill:#fff9c4
630633
style KVBM fill:#fff9c4
631-
style SL fill:#f5f5f5
632-
```
634+
style SL fill:#f5f5f5`}
635+
</ZoomableMermaid>
633636

634637
**Architecture Layers:**
635638

website/docs/proposals/prompt-classification-routing.md

Lines changed: 6 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -121,8 +121,10 @@ embedding_similarity:
121121
122122
### High-Level System Design
123123
124-
```mermaid
125-
graph TD
124+
import ZoomableMermaid from '@site/src/components/ZoomableMermaid';
125+
126+
<ZoomableMermaid title="System Architecture Overview" defaultZoom={5.5}>
127+
{`graph TD
126128
A[Envoy External Processor<br/>semantic-router ExtProc] --> B[Request Handler<br/>handleModelRouting]
127129

128130
B --> C{Execution Path}
@@ -166,8 +168,8 @@ graph TD
166168
style E2 fill:#fff9c4
167169
style F fill:#c8e6c9
168170
style H fill:#ffcdd2
169-
style M fill:#c8e6c9
170-
```
171+
style M fill:#c8e6c9`}
172+
</ZoomableMermaid>
171173

172174
### Component Breakdown
173175

0 commit comments

Comments
 (0)