Support for Anthropic added

README.md

@@ -67,9 +67,10 @@ The author(s) are looking to add core maintainers for this opensource project. R
     -   [How to request more loaders](#more-loaders-coming-soon)
 -   [LLMs](#llms)
     -   [OpenAI](#openai)
+    -   [Azure OpenAI](#azure-openai)
     -   [Mistral](#mistral)
     -   [Hugging Face](#hugging-face)
-    -   [Azure OpenAI](#azure-openai)
+    -   [Anthropic](#anthropic)
     -   [Bring your own LLMs](#use-custom-llm-model)
     -   [Request support for new LLMs](#more-llms-coming-soon)
 -   [Embedding Models](#embedding-models)
@@ -358,6 +359,36 @@ const ragApplication = await new RAGApplicationBuilder()
 
 **Note:** GPT 3.5 Turbo is used as the default model if you do not specifiy one.
 
+## Azure OpenAI
+
+In order to be able to use an OpenAI model on Azure, it first needs to be deployed. Please refer to [Azure OpenAI documentation](https://learn.microsoft.com/en-us/azure/cognitive-services/openai/) on how to deploy a model on Azure. To run this library, you will need to deploy two models -
+
+-   text-embedding-ada
+-   GPT-3.5-turbo (or the 4 series)
+
+Once these models are deployed, using Azure OpenAI instead of the regular OpenAI is easy to do. Just follow these steps -
+
+-   Remove the `OPENAI_API_KEY` environment variable if you have set it already.
+
+-   Set the following environment variables -
+
+```bash
+# Set this to `azure`
+OPENAI_API_TYPE=azure
+# The API version you want to use
+AZURE_OPENAI_API_VERSION=2023-05-15
+# The base URL for your Azure OpenAI resource.  You can find this in the Azure portal under your Azure OpenAI resource.
+export AZURE_OPENAI_BASE_PATH=https://your-resource-name.openai.azure.com/openai/deployments
+# The API key1 or key2 for your Azure OpenAI resource
+export AZURE_OPENAI_API_KEY=<Your Azure OpenAI API key>
+# The deployment name you used for your embedding model
+AZURE_OPENAI_API_EMBEDDINGS_DEPLOYMENT_NAME=text-embedding-ada-002
+# The deployment name you used for your llm
+AZURE_OPENAI_API_DEPLOYMENT_NAME=gpt-35-turbo
+```
+
+You can all set and can now run the Azure OpenAI LLMs using the [`OpenAi` model](#openai) steps detailed above.
+
 ## Mistral
 
 To use Mirstal's models, you will need to get an API Key from Mistral. You can do this from their [console](https://console.mistral.ai/user/api-keys/). Once you have obtained a key, set Mistral as your LLM of choice -
@@ -397,35 +428,29 @@ const ragApplication = await new RAGApplicationBuilder()
 
 To use these 'not-free' models via HuggingFace, you need to subscribe to their [Pro plan](https://huggingface.co/pricing) or create a custom [inference endpoint](https://ui.endpoints.huggingface.co/). It is possible to self host these models for free and run them locally via Ollama - support for which is coming soon.
 
-## Azure OpenAI
+## Anthropic
 
-In order to be able to use an OpenAI model on Azure, it first needs to be deployed. Please refer to [Azure OpenAI documentation](https://learn.microsoft.com/en-us/azure/cognitive-services/openai/) on how to deploy a model on Azure. To run this library, you will need to deploy two models -
+To use Anthropic's Claude models, you will need to get an API Key from Anthropic. You can do this from their [console](https://console.anthropic.com/settings/keys). Once you obtain a key, set it in the environment variable, like so -
 
--   text-embedding-ada
--   GPT-3.5-turbo (or the 4 series)
+```bash
+ANTHROPIC_API_KEY="<Your key>"
+```
 
-Once these models are deployed, using Azure OpenAI instead of the regular OpenAI is easy to do. Just follow these steps -
+Once this is done, it is relatively easy to use Anthropic's Claude in your RAG application. Simply set Anthropic as your LLM of choice -
 
--   Remove the `OPENAI_API_KEY` environment variable if you have set it already.
+```TS
+const ragApplication = await new RAGApplicationBuilder()
+.setModel(new Anthropic())
+```
 
--   Set the following environment variables -
+By default, the `claude-3-sonnet-20240229` model from Anthropic is used. If you want to use a different Anthropic model, you can specify it via the optional parameter to the Anthropic constructor, like so -
 
-```bash
-# Set this to `azure`
-OPENAI_API_TYPE=azure
-# The API version you want to use
-AZURE_OPENAI_API_VERSION=2023-05-15
-# The base URL for your Azure OpenAI resource.  You can find this in the Azure portal under your Azure OpenAI resource.
-export AZURE_OPENAI_BASE_PATH=https://your-resource-name.openai.azure.com/openai/deployments
-# The API key1 or key2 for your Azure OpenAI resource
-export AZURE_OPENAI_API_KEY=<Your Azure OpenAI API key>
-# The deployment name you used for your embedding model
-AZURE_OPENAI_API_EMBEDDINGS_DEPLOYMENT_NAME=text-embedding-ada-002
-# The deployment name you used for your llm
-AZURE_OPENAI_API_DEPLOYMENT_NAME=gpt-35-turbo
+```TS
+const ragApplication = await new RAGApplicationBuilder()
+.setModel(new Anthropic({ modelName: "..." }))
 ```
 
-You can now run the Azure OpenAI LLMs using the [`OpenAi` model](#openai) detailed above.
+You can read more about the various models provided by Anthropic [here](https://docs.anthropic.com/claude/docs/models-overview).
 
 ## Use custom LLM model
 

package.json

@@ -1,6 +1,6 @@
 {
     "name": "@llm-tools/embedjs",
-    "version": "0.0.68",
+    "version": "0.0.71",
     "description": "A NodeJS RAG framework to easily work with LLMs and custom datasets",
     "main": "dist/index.js",
     "types": "dist/index.d.ts",
@@ -24,6 +24,9 @@
         "llm",
         "gpt",
         "openai",
+        "anthropic",
+        "claude",
+        "qdrant",
         "chatgpt",
         "hugging-face",
         "mistral",
@@ -49,16 +52,17 @@
     "homepage": "https://github.com/llm-tools/embedjs#readme",
     "dependencies": {
         "@huggingface/inference": "^2.6.7",
+        "@langchain/anthropic": "^0.1.16",
         "@langchain/cohere": "^0.0.8",
-        "@langchain/community": "^0.0.51",
-        "@langchain/core": "^0.1.59",
+        "@langchain/community": "^0.0.52",
+        "@langchain/core": "^0.1.60",
         "@langchain/mistralai": "^0.0.19",
         "@langchain/openai": "^0.0.28",
         "axios": "^1.6.8",
         "confluence.js": "^1.7.4",
         "debug": "^4.3.4",
         "html-to-text": "^9.0.5",
-        "langchain": "^0.1.35",
+        "langchain": "^0.1.36",
         "md5": "^2.3.0",
         "pdf-parse-fork": "^1.2.0",
         "sitemapper": "^3.1.8",

src/core/rag-application.ts

@@ -67,7 +67,7 @@ export class RAGApplication {
                 await this.addLoader(loader);
             }
         }
-        this.debug('Initial loaders added');
+        this.debug('Initialized pre-loaders');
     }
 
     private async batchLoadEmbeddings(loaderUniqueId: string, formattedChunks: Chunk[]) {

src/index.ts

@@ -22,6 +22,7 @@ import { OpenAi3LargeEmbeddings } from './embeddings/openai-3large-embeddings.js
 import { OpenAi3SmallEmbeddings } from './embeddings/openai-3small-embeddings.js';
 import { Mistral } from './models/mistral-model.js';
 import { HuggingFace } from './models/huggingface-model.js';
+import { Anthropic } from './models/anthropic-model.js';
 
 export {
     RAGApplication,
@@ -48,4 +49,5 @@ export {
     OpenAi3SmallEmbeddings,
     Mistral,
     HuggingFace,
+    Anthropic,
 };