Add support for gpt-4-turbo 128K model (#340)

resolves #336
Add support for gpt-4-turbo 128K model

README.md

@@ -73,7 +73,7 @@ This monorepo consists of three main sections:
 ### Requirements
 - `yarn` and `node` on your machine
 - `python` 3.9+ for running scripts in `collector/`.
-- access to an LLM like `GPT-3.5`, `GPT-4`.
+- access to an LLM like `GPT-3.5`, `GPT-4`, etc.
 - (optional) a vector database like Pinecone, qDrant, Weaviate, or Chroma*.
 *AnythingLLM by default uses a built-in vector db called LanceDB.
 

frontend/src/components/LLMSelection/AzureAiOptions/index.jsx

@@ -63,6 +63,7 @@ export default function AzureAiOptions({ settings }) {
           <option value={16384}>16,384 (gpt-3.5-16k)</option>
           <option value={8192}>8,192 (gpt-4)</option>
           <option value={32768}>32,768 (gpt-4-32k)</option>
+          <option value={128000}>128,000 (gpt-4-turbo)</option>
         </select>
       </div>
 

frontend/src/components/LLMSelection/OpenAiOptions/index.jsx

@@ -84,17 +84,19 @@ function OpenAIModelSelection({ apiKey, settings }) {
         className="bg-zinc-900 border border-gray-500 text-white text-sm rounded-lg block w-full p-2.5"
       >
         <optgroup label="General LLM models">
-          {["gpt-3.5-turbo", "gpt-4"].map((model) => {
-            return (
-              <option
-                key={model}
-                value={model}
-                selected={settings.OpenAiModelPref === model}
-              >
-                {model}
-              </option>
-            );
-          })}
+          {["gpt-3.5-turbo", "gpt-4", "gpt-4-1106-preview", "gpt-4-32k"].map(
+            (model) => {
+              return (
+                <option
+                  key={model}
+                  value={model}
+                  selected={settings.OpenAiModelPref === model}
+                >
+                  {model}
+                </option>
+              );
+            }
+          )}
         </optgroup>
         {customModels.length > 0 && (
           <optgroup label="Your fine-tuned models">

server/utils/AiProviders/openAi/index.js

@@ -25,13 +25,22 @@ class OpenAiLLM extends OpenAiEmbedder {
         return 4096;
       case "gpt-4":
         return 8192;
+      case "gpt-4-1106-preview":
+        return 128000;
+      case "gpt-4-32k":
+        return 32000;
       default:
         return 4096; // assume a fine-tune 3.5
     }
   }
 
   async isValidChatCompletionModel(modelName = "") {
-    const validModels = ["gpt-4", "gpt-3.5-turbo"];
+    const validModels = [
+      "gpt-4",
+      "gpt-3.5-turbo",
+      "gpt-4-1106-preview",
+      "gpt-4-32k",
+    ];
     const isPreset = validModels.some((model) => modelName === model);
     if (isPreset) return true;