adds preliminary gpt4o

package-lock.json

@@ -5445,9 +5445,9 @@
       }
     },
     "node_modules/tiktoken": {
-      "version": "1.0.10",
-      "resolved": "https://registry.npmjs.org/tiktoken/-/tiktoken-1.0.10.tgz",
-      "integrity": "sha512-gF8ndTCNu7WcRFbl1UUWaFIB4CTXmHzS3tRYdyUYF7x3C6YR6Evoao4zhKDmWIwv2PzNbzoQMV8Pxt+17lEDbA=="
+      "version": "1.0.14",
+      "resolved": "https://registry.npmjs.org/tiktoken/-/tiktoken-1.0.14.tgz",
+      "integrity": "sha512-g5zd5r/DoH8Kw0fiYbYpVhb6WO8BHO1unXqmBBWKwoT17HwSounnDtMDFUKm2Pko8U47sjQarOe+9aUrnqmmTg=="
     },
     "node_modules/to-regex-range": {
       "version": "5.0.1",

src/proxy/middleware/request/preprocessors/validate-context-size.ts

@@ -54,6 +54,8 @@ export const validateContextSize: RequestPreprocessor = async (req) => {
   let modelMax: number;
   if (model.match(/gpt-3.5-turbo-16k/)) {
     modelMax = 16384;
+  } else if (model.match(/^gpt-4o/)) {
+    modelMax = 128000;
   } else if (model.match(/gpt-4-turbo(-\d{4}-\d{2}-\d{2})?$/)) {
     modelMax = 131072;
   } else if (model.match(/gpt-4-turbo(-preview)?$/)) {

src/proxy/openai.ts

@@ -28,6 +28,8 @@ import {
 
 // https://platform.openai.com/docs/models/overview
 export const KNOWN_OPENAI_MODELS = [
+  "gpt-4o",
+  "gpt-4o-2024-05-13",
   "gpt-4-turbo", // alias for latest gpt4-turbo stable
   "gpt-4-turbo-2024-04-09", // gpt4-turbo stable, with vision
   "gpt-4-turbo-preview", // alias for latest turbo preview

src/shared/models.ts

@@ -75,6 +75,7 @@ export const LLM_SERVICES = (<A extends readonly LLMService[]>(
 ] as const);
 
 export const OPENAI_MODEL_FAMILY_MAP: { [regex: string]: OpenAIModelFamily } = {
+  "^gpt-4o": "gpt4-turbo", //todo: separate model family
   "^gpt-4-turbo(-\\d{4}-\\d{2}-\\d{2})?$": "gpt4-turbo",
   "^gpt-4-turbo(-preview)?$": "gpt4-turbo",
   "^gpt-4-(0125|1106)(-preview)?$": "gpt4-turbo",

src/shared/stats.ts

@@ -6,6 +6,10 @@ import { ModelFamily } from "./models";
 export function getTokenCostUsd(model: ModelFamily, tokens: number) {
   let cost = 0;
   switch (model) {
+    // case "gpt4o":
+    // case "azure-gpt4o":
+    //   cost = 0.000005;
+    //   break;
     case "azure-gpt4-turbo":
     case "gpt4-turbo":
       cost = 0.00001;