Merge branch 'main' into add/model-list

README.md

@@ -70,7 +70,7 @@ Jan is an open-source ChatGPT alternative that runs 100% offline on your compute
   <tr style="text-align: center">
     <td style="text-align:center"><b>Experimental (Nighlty Build)</b></td>
     <td style="text-align:center" colspan="4">
-      <a href='https://github.com/janhq/jan/actions/runs/7328809734'>
+      <a href='https://github.com/janhq/jan/actions/runs/7329755769'>
         <b>Github action artifactory</b>
       </a>
     </td>

docs/docs/guides/04-using-models/02-import-manually.mdx

@@ -109,6 +109,8 @@ Copy the following configurations into the `model.json`.
 2. Make sure the `source_url` property is the direct binary download link ending in `.gguf`. In HuggingFace, you can find the directl links in `Files and versions` tab.
 3. Ensure you are using the correct `prompt_template`. This is usually provided in the HuggingFace model's description page.
 
+> Note: Currently, the filename must be `model.json` and the ID has to be equal to the foldername. In the `model.json`, you have to include the `state` property and set it to `ready` for Jan to recognize the model.
+
 ```js
 {
     "source_url": "https://huggingface.co/janhq/trinity-v1-GGUF/resolve/main/trinity-v1.Q4_K_M.gguf",
@@ -130,6 +132,7 @@ Copy the following configurations into the `model.json`.
       "tags": ["7B", "Merged", "Featured"],
       "size": 4370000000
     },
+    "state": "ready",
     "engine": "nitro"
   }
 ```

extensions/inference-nitro-extension/src/module.ts

@@ -68,6 +68,15 @@ async function loadModel(nitroResourceProbe: any | undefined) {
   if (!nitroResourceProbe) nitroResourceProbe = await getResourcesInfo();
   return killSubprocess()
     .then(() => tcpPortUsed.waitUntilFree(PORT, 300, 5000))
+    // wait for 500ms to make sure the port is free for windows platform
+    .then(() => {
+      if (process.platform === "win32") {
+        return sleep(500);
+      }
+      else {
+        return sleep(0);
+      }
+    })
     .then(() => spawnNitroProcess(nitroResourceProbe))
     .then(() => loadLLMModel(currentSettings))
     .then(validateModelStatus)
@@ -78,6 +87,11 @@ async function loadModel(nitroResourceProbe: any | undefined) {
     });
 }
 
+// Add function sleep
+function sleep(ms) {
+  return new Promise((resolve) => setTimeout(resolve, ms));
+}
+
 function promptTemplateConverter(promptTemplate) {
   // Split the string using the markers
   const systemMarker = "{system_message}";