x

langchain-ai · Dec 16, 2024 · e50d15b · e50d15b
1 parent 5eb1098
commit e50d15b
Showing 1 changed file with 25 additions and 28 deletions.
diff --git a/docs/docs/integrations/text_embedding/localai.ipynb b/docs/docs/integrations/text_embedding/localai.ipynb
@@ -8,22 +8,40 @@
    "source": [
     "# LocalAI\n",
     "\n",
+    ":::info\n",
+    "\n",
+    "`langchain-localai` is a 3rd party integration package for LocalAI. It provides a simple way to use LocalAI services in Langchain.\n",
+    "\n",
+    "The source code is available on [Github](https://github.com/mkhludnev/langchain-localai)\n",
+    "\n",
+    ":::\n",
+    "\n",
     "Let's load the LocalAI Embedding class. In order to use the LocalAI Embedding class, you need to have the LocalAI service hosted somewhere and configure the embedding models. See the documentation at https://localai.io/basics/getting_started/index.html and https://localai.io/features/embeddings/index.html."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
-   "id": "0be1af71",
+   "execution_count": null,
+   "id": "799d1f77",
    "metadata": {},
    "outputs": [],
    "source": [
-    "from langchain_community.embeddings import LocalAIEmbeddings"
+    "%pip install -U langchain-localai"
    ]
   },
   {
    "cell_type": "code",
    "execution_count": 2,
+   "id": "0be1af71",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain_localai import LocalAIEmbeddings"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
    "id": "2c66e5da",
    "metadata": {},
    "outputs": [],
@@ -35,7 +53,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "id": "01370375",
    "metadata": {},
    "outputs": [],
@@ -45,7 +63,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "id": "bfb6142c",
    "metadata": {},
    "outputs": [],
@@ -136,27 +154,11 @@
     "# if you are behind an explicit proxy, you can use the OPENAI_PROXY environment variable to pass through\n",
     "os.environ[\"OPENAI_PROXY\"] = \"http://proxy.yourcompany.com:8080\""
    ]
-  },
-  {
-   "cell_type": "markdown",
-   "id": "4a824a90",
-   "metadata": {},
-   "source": [
-    "## Limitations \n",
-    "\n",
-    "`langchain_community.embeddings.LocalAIEmbeddings` has two issues: \n",
-    " - it depends on Open AI SDK v0, which is outdated\n",
-    " - and it requests document embeddings one by one in `embed_documents` without bulking them into the single request. \n",
-    "\n",
-    " `langchain-localai` is the separate integration package provided for resolving these issues: \n",
-    " - the distribution: https://pypi.org/project/langchain-localai/ \n",
-    " - the source code: https://github.com/mkhludnev/langchain-localai"
-   ]
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": ".venv",
    "language": "python",
    "name": "python3"
   },
@@ -170,12 +172,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
-  },
-  "vscode": {
-   "interpreter": {
-    "hash": "e971737741ff4ec9aff7dc6155a1060a59a8a6d52c757dbbe66bf8ee389494b1"
-   }
+   "version": "3.11.4"
   }
  },
  "nbformat": 4,