EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[doc] update router document (#11767)

Browse Source
This commit is contained in:
Keyang Ru
2025-10-17 10:26:54 -07:00
committed by GitHub
parent a5978a20f0
commit 2bc3fcd420
2 changed files with 20 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
docs/advanced_features/router.md
View File

@@ -81,7 +81,7 @@ Comprehensive example:
python3 -m sglang_router.launch_server \
--host 0.0.0.0 \
--port 8080 \
--model /raid/models/meta-llama/Llama-3.1-8B-Instruct \
--model meta-llama/Llama-3.1-8B-Instruct \
--tp-size 1 \
--dp-size 8 \
--grpc-mode \
@@ -91,7 +91,7 @@ python3 -m sglang_router.launch_server \
--router-health-success-threshold 2 \
--router-health-check-timeout-secs 6000 \
--router-health-check-interval-secs 60 \
--router-model-path /raid/models/meta-llama/Llama-3.1-8B-Instruct \
--router-model-path meta-llama/Llama-3.1-8B-Instruct \
--router-policy round_robin \
--router-log-level debug
```
@@ -117,7 +117,7 @@ Use SRT gRPC workers to unlock the highest throughput and access native reasonin
```bash
# Workers expose gRPC endpoints
python -m sglang.launch_server \
--model /raid/models/meta-llama/Llama-3.1-8B-Instruct \
--model meta-llama/Llama-3.1-8B-Instruct \
--grpc-mode \
--port 20000
@@ -152,7 +152,6 @@ Proxy OpenAI-compatible endpoints (OpenAI, xAI, etc.) while keeping history and
python -m sglang_router.launch_router \
--backend openai \
--worker-urls https://api.openai.com \
--api-key "$OPENAI_API_KEY" \
--history-backend memory
```
@@ -171,7 +170,7 @@ curl -X POST http://localhost:30000/workers \
-d '{"url":"grpc://0.0.0.0:31000","worker_type":"regular"}'
# Inspect registry
curl http://localhost:30000/workers | jq
curl http://localhost:30000/workers
# Remove a worker
curl -X DELETE http://localhost:30000/workers/grpc://0.0.0.0:31000
@@ -278,8 +277,18 @@ PD deployments can specify `--prefill-selector` and `--decode-selector` plus the
| `oracle` | Oracle Autonomous Database-backed storage (pooled connections). | `--history-backend oracle` |
Oracle configuration (choose DSN *or* TNS alias):
Install the Oracle Instant Client and set `LD_LIBRARY_PATH` accordingly.
Choose **one** connection method:
```bash
# Option 1: Full connection descriptor
export ATP_DSN="(description=(address=(protocol=tcps)(port=1522)(host=adb.region.oraclecloud.com))(connect_data=(service_name=service_name)))"
# Option 2: TNS alias (requires wallet)
export ATP_TNS_ALIAS="sglroutertestatp_high"
export ATP_WALLET_PATH="/path/to/wallet"
```
Provide database credentials and optional pool sizing:
```bash
export ATP_DSN="tcps://host:port/service" # or use ATP_TNS_ALIAS + ATP_WALLET_PATH
export ATP_USER="admin"
export ATP_PASSWORD="secret"
export ATP_POOL_MIN=4
@@ -320,7 +329,6 @@ Use CLI flags to select parsers:
| `POST` | `/v1/completions` | OpenAI-compatible text completions. |
| `POST` | `/v1/responses` | Create background responses (agentic loops). |
| `GET` | `/v1/responses/{id}` | Retrieve stored responses. |
| `GET` | `/v1/responses/{id}/input` | List captured input items. |
| `POST` | `/v1/embeddings` | Forward embedding requests. |
| `POST` | `/v1/rerank` | Ranking endpoint (`/rerank` synonym). |
| `POST` | `/v1/conversations` | Create conversation metadata. |

15
sgl-router/README.md
View File

@@ -147,7 +147,7 @@ curl -X POST http://localhost:30000/workers \
}'
# Inspect registered workers
curl http://localhost:30000/workers | jq
curl http://localhost:30000/workers
```
Sample response (http workers):
```json
@@ -194,13 +194,11 @@ Route requests to OpenAI or OpenAI-compatible endpoints:
python3 -m sglang_router.launch_router \
--backend openai \
--worker-urls https://api.openai.com \
--api-key "$OPENAI_API_KEY"
# Route to custom OpenAI-compatible endpoint (Gemini, xAI, etc.)
python3 -m sglang_router.launch_router \
--backend openai \
--worker-urls http://my-openai-compatible-service:8000 \
--api-key "tenant-api-key"
```
**Notes**
@@ -218,7 +216,7 @@ Add flags as needed for production deployments:
python3 -m sglang_router.launch_server \
--host 0.0.0.0 \
--port 8080 \
--model /raid/models/meta-llama/Llama-3.1-8B-Instruct \
--model meta-llama/Llama-3.1-8B-Instruct \
--tp-size 1 \
--dp-size 8 \
--grpc-mode
@@ -240,7 +238,7 @@ Use upstream SGLang binaries to start dedicated worker processes.
- **Prefill worker server (gRPC mode)**:
```bash
python3 -m sglang.launch_server \
--model /raid/models/meta-llama/Llama-3.1-8B-Instruct \
--model meta-llama/Llama-3.1-8B-Instruct \
--port 20000 \
--tp-size 1 \
--grpc-mode
@@ -312,7 +310,7 @@ The HTTP router exposes the full OpenAI-compatible surface area (`/generate`, `/
### OpenAI Router
- Proxies OpenAI-compatible chat completions and responses APIs, preserving headers and SSE streams end-to-end.
- Supports `/v1/responses` background jobs with cancellation, deletion, and listing input items—enabling agentic, multi-turn orchestration without persisting data at remote vendor endpoints.
- Conversation APIs (`/v1/conversations` and `/items`) interact with the configured conversation storage backend for compliant chat-history management. Conversation state lives at the router tier, so the same history can drive different models or MCP loops without leaking data to upstream vendors.
- Conversation APIs (`/v1/conversations` and `/v1/conversations/{id}/items`) interact with the configured conversation storage backend for compliant chat-history management. Conversation state lives at the router tier, so the same history can drive different models or MCP loops without leaking data to upstream vendors.
- Chat history, agentic multi-turn `/v1/responses`, and the native MCP client (STDIO/HTTP/SSE/Streamable transports) are designed to satisfy enterprise data-privacy requirements by keeping sensitive state within the router.
### Request Endpoints
@@ -323,10 +321,7 @@ The HTTP router exposes the full OpenAI-compatible surface area (`/generate`, `/
| `POST /v1/completions` | OpenAI-compatible text completions. |
| `POST /v1/responses` | Create background responses, returns response IDs. |
| `GET /v1/responses/{id}` | Retrieve stored responses. |
| `POST /v1/responses/{id}/cancel` | Cancel in-flight background jobs. |
| `DELETE /v1/responses/{id}` | Delete stored response. |
| `GET /v1/responses/{id}/input` | List captured input items. |
| Conversation endpoints (`/v1/conversations`, `/v1/conversations/{id}`, `/items`) | Manage chat history. |
| Conversation endpoints (`/v1/conversations`, `/v1/conversations/{id}`, `/v1/conversations/{id}/items`) | Manage chat history. |
| `POST /v1/embeddings` | Forward embedding requests. |
| `POST /v1/rerank`, `POST /rerank` | Ranking APIs. |