feat: add vLLM engine with on-request model switching

README.md

@@ -38,7 +38,7 @@ one, and both report live GPU and memory use throughout.
 | **Architectures** | x64 and arm64 on all three. Windows on ARM is experimental. |
 | **Installers** | Windows `.exe`; Linux `.deb`; macOS `.dmg`. On other Linux distributions, [build from source](docs/building.mdx). |
 | **Mixing nodes** | Windows, Linux, and macOS nodes can all be paired with each other |
-| **Inference engines** | Ollama, LM Studio, and llama.cpp |
+| **Inference engines** | Ollama, LM Studio, llama.cpp, and vLLM (Linux with an NVIDIA GPU, or Apple Silicon macOS) |
 
 **PAIR running on a machine does not mean an engine will.** PAIR itself runs on
 any supported Windows, Linux, or macOS machine. Each engine sets its own requirements
@@ -100,7 +100,7 @@ you want by its full filename instead.
   **Settings → Service** and read the status there.
 
 - **Get an engine running.** On the node's card, open **Engine settings** and
-  select **Install** next to Ollama, LM Studio, or llama.cpp. PAIR downloads and
+  select **Install** next to Ollama, LM Studio, llama.cpp, or (on Linux or Apple Silicon macOS) vLLM. PAIR downloads and
   sets the engine up for you, so nothing needs to be in place beforehand.
 
   ![The Install engines dialog with Ollama downloading, reporting progress as it installs.](docs/assets/onboarding/engine-lifecycle/01-engine-installing.png)
@@ -199,7 +199,7 @@ Your data means settings, logs, cluster identity and certificates, and any engin
 PAIR installed for you. An engine the uninstaller cannot remove — one still
 holding a file open, say — is left in place rather than stopping the uninstall;
 remove it with its own uninstaller. **Model weights are not touched** — they live in the
-engine's own storage (`~/.ollama`, `~/.llamacpp`, `~/.lmstudio/models`), so
+engine's own storage (`~/.ollama`, `~/.llamacpp`, `~/.lmstudio/models`, `~/.cache/huggingface`), so
 removing PAIR does not delete the models you downloaded. Delete those through the
 engine, or by removing its directory. An engine you installed yourself is also
 left alone.
@@ -284,7 +284,7 @@ feedback and contributions will help shape priorities.
 ### Engines and integrations
 
 - [x] llama.cpp support.
-- [ ] vLLM support.
+- [x] vLLM support.
 - [ ] EXO support.
 - [ ] ComfyUI integration.
 - [ ] Unsloth support.

desktop/docs/architecture.md

@@ -250,7 +250,7 @@ not select or pin proxy routes.
 An NVPAIR-launched engine binds to loopback and is never directly LAN-reachable.
 Peers reach it only through the node's proxy over a cluster-mTLS ingress, so
 discovery advertises the promoted proxy port for `ol`/`lm` rather than the
-engine's private port; llama.cpp uses the additional `lc` key. Transport
+engine's private port; llama.cpp uses the additional `lc` key and vLLM the `vl` key. Transport
 security is backend-owned; Electron only
 reflects the advertised proxy port and reads a remote engine's real port from
 `engine:remote-get-installed` facts.
@@ -408,7 +408,7 @@ deletes finish. Unpackaged (`electron-vite dev`) builds omit relaunch and the UI
 tells the developer to run `npm start` again — quitting Electron tears down the Vite
 renderer server, so auto-relaunch would come up with nothing to load.
 
-The engines' model stores (`~/.ollama`, `~/.llamacpp`, `~/.lmstudio/models`) and
+The engines' model stores (`~/.ollama`, `~/.llamacpp`, `~/.lmstudio/models`, `~/.cache/huggingface`) and
 the application install tree are never removed. The reset does uninstall the
 engines PAIR installed, through `engine:uninstall-managed` while the broker is
 still running, so an engine a vendor installer placed in the user's home is

desktop/docs/frontend-api.md

@@ -99,11 +99,19 @@ They do not imply a WebSocket connection. Browser clients are not supported.
 - `onSettingsDisconnected(callback)` reports that a node's settings authority
   became unreachable, so its cached snapshot is stale.
 
-`EngineType` currently includes `ollama`, `lm-studio`, and `llama-cpp`.
+`EngineType` currently includes `ollama`, `lm-studio`, `llama-cpp`, and `vllm`.
+vLLM serves one model per process, so `unloadModel` (Eject) stops the engine,
+which frees its GPU memory, and `loadModel` on a stopped engine starts it. Its model
+list is the Hugging Face cache, which stays available while the engine is stopped, `deleteModel` removes a cached model (not the one
+loaded in vLLM; deleting an ejected or stopped vLLM's model leaves it with no model
+selected until `loadModel` picks one), `pullModel` downloads a Hugging Face repo into
+that cache (no running engine needed), and `loadModel` on a cached model switches
+the served model by restarting vLLM on it.
 `searchHub(engineType, query?)` serves Ollama's locked local catalog and cached
-live Hugging Face catalogs for LM Studio and llama.cpp. A non-empty llama.cpp
-query performs an explicit bounded search across public Hugging Face GGUF
-repositories; other calls return the populated engine catalog. llama.cpp pull
+live Hugging Face catalogs for LM Studio, llama.cpp and vLLM. A non-empty
+llama.cpp or vLLM query performs an explicit bounded search across public
+Hugging Face repositories (GGUF for llama.cpp, safetensors chat models for vLLM);
+other calls return the populated engine catalog. llama.cpp pull
 IDs are exact `owner/repository:quantization` values; load, unload, and delete
 all use that same exact id.
 

desktop/docs/services-api.md

@@ -43,6 +43,7 @@
 - ⚠️ nvpair-proxy → node/set-local-backend
 - ⚠️ nvpair-ui-broker → discovery:unsubscribe
 - ⚠️ nvpair-ui-broker → engine:configure-launch
+- ⚠️ nvpair-ui-broker → engine:restart
 - ⚠️ nvpair-ui-broker → engine:set-port
 - ⚠️ nvpair-ui-broker → engine:set-reserved-port
 - ⚠️ nvpair-ui-broker → engine:unsubscribe
@@ -270,9 +271,11 @@
 | `engine:get-settings` | request (we call) | ✅ yes |
 | `engine:install` | request (we call) | ✅ yes |
 | `engine:preview-settings` | request (we call) | ✅ yes |
+| `engine:restart` | request (we call) | ⚠️ not called |
 | `engine:set-port` | request (we call) | ⚠️ not called |
 | `engine:set-reserved-port` | request (we call) | ⚠️ not called |
 | `engine:start` | request (we call) | ✅ yes |
+| `engine:stop` | request (we call) | ✅ yes |
 | `engine:subscribe` | request (we call) | ✅ yes |
 | `engine:unsubscribe` | request (we call) | ⚠️ not called |
 | `errors:get-initial` | request (we call) | ✅ yes |

desktop/docs/services-backend.md

@@ -116,7 +116,7 @@ reserved for inference clients.
 | ---------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------- |
 | `app:ready`                                          | Complete broker startup and refresh snapshots                                                                                   | `state:request-refresh`                                   |
 | `discovery:nodes-changed`                            | Replace discovery snapshot and diff nodes                                                                                       | `discovery:nodes-changed`, `nodes:upsert`, `nodes:remove` |
-| `ollama-proxy:ready` / `lmstudio-proxy:ready` / `llamacpp-proxy:ready` | Record engine proxy port                                                                                          | `engines:state-changed`                                   |
+| `ollama-proxy:ready` / `lmstudio-proxy:ready` / `llamacpp-proxy:ready` / `vllm-proxy:ready` | Record engine proxy port                                                                                          | `engines:state-changed`                                   |
 | proxy `node/*`                                       | Update per-engine node presence; the advertised port is the peer's promoted proxy port (not the engine's private loopback port) | node and engine pushes                                    |
 | `engine:ready` / `engine:state-changed`              | Update engine facts and models                                                                                                  | `engines:state-changed`                                   |
 | `engine:settings-changed`                            | Validate and republish the owning node's settings snapshot                                                                      | `engines:settings-changed`                                |