feat: Model Updates

Author: samestrin

README.md

@@ -123,7 +123,7 @@ Passing a more complex message object is just as simple. The same rules apply:
 
 ```javascript
 const message = {
-  model: 'gpt-3.5-turbo',
+  model: 'gpt-4o-mini',
   messages: [
     { role: 'system', content: 'You are a helpful assistant.' },
     { role: 'user', content: 'Explain the importance of low latency LLMs.' },

docs/models.md

@@ -63,7 +63,7 @@ const geminiResult = await LLMInterface.sendMessage("gemini", "Explain the impor
 Changing the aliases is easy:
 
 ```javascript
-LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
+LLMInterface.setModelAlias("openai", "default", "gpt-4o-mini");
 ```
 
 ## Model Alias Values
@@ -197,7 +197,7 @@ LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
 - `default`: forefront/Mistral-7B-Instruct-v0.2-chatml
 - `large`: forefront/Mistral-7B-Instruct-v0.2-chatml
 - `small`: forefront/Mistral-7B-Instruct-v0.2-chatml
-- `agent`: 
+- `agent`:
 
 
 ![](https://samestrin.github.io/media/llm-interface/icons/blank.png)
@@ -318,7 +318,7 @@ LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
 - `default`: Neets-7B
 - `large`: mistralai/Mixtral-8X7B-Instruct-v0.1
 - `small`: Neets-7B
-- `agent`: 
+- `agent`:
 
 
 ![](https://samestrin.github.io/media/llm-interface/icons/blank.png)
@@ -362,17 +362,15 @@ LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
 - `default`: llama3
 - `large`: llama3
 - `small`: llama3
-- `agent`: 
+- `agent`:
 
 
 ![](https://samestrin.github.io/media/llm-interface/icons/blank.png)
 ### [OpenAI](providers/openai.md)
 
-
-
-- `default`: gpt-3.5-turbo
+- `default`: gpt-4o-mini
 - `large`: gpt-4o
-- `small`: gpt-3.5-turbo
+- `small`: gpt-4o-mini
 - `agent`: gpt-4o
 
 
@@ -381,10 +379,10 @@ LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
 
 ![perplexity](https://samestrin.github.io/media/llm-interface/icons/perplexity.png)
 
-- `default`: llama-3-sonar-large-32k-online
-- `large`: llama-3-sonar-large-32k-online
-- `small`: llama-3-sonar-small-32k-online
-- `agent`: llama-3-sonar-large-32k-online
+- `default`: llama-3.1-sonar-large-128k-chat
+- `large`: llama-3.1-sonar-large-128k-chat
+- `small`: llama-3.1-sonar-small-128k-chat
+- `agent`: llama-3.1-70b-instruct
 
 
 ![](https://samestrin.github.io/media/llm-interface/icons/blank.png)
@@ -461,7 +459,7 @@ LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
 - `default`: palmyra-x-002-32k
 - `large`: palmyra-x-002-32k
 - `small`: palmyra-x-002-32k
-- `agent`: 
+- `agent`:
 
 
 ![](https://samestrin.github.io/media/llm-interface/icons/blank.png)
@@ -473,4 +471,3 @@ LLMInterface.setModelAlias("openai", "default", "gpt-3.5-turbo");
 - `large`: glm-4
 - `small`: glm-4-flash
 - `agent`: glm-4
-

docs/providers/perplexity.md

@@ -30,12 +30,12 @@ main();
 
 ### Model Aliases
 
-The following model aliases are provided for this provider. 
+The following model aliases are provided for this provider.
 
-- `default`: llama-3-sonar-large-32k-online
-- `large`: llama-3-sonar-large-32k-online
-- `small`: llama-3-sonar-small-32k-online
-- `agent`: llama-3-sonar-large-32k-online
+- `default`: llama-3.1-sonar-large-128k-chat
+- `large`: llama-3.1-sonar-large-128k-chat
+- `small`: llama-3.1-sonar-small-128k-chat
+- `agent`: llama-3.1-8b-instruct
 
 
 ## Options

src/config/providers/openai.json

@@ -1 +1,20 @@
-{"url":"https://api.openai.com/v1/chat/completions","model":{"default":"gpt-3.5-turbo","large":"gpt-4o","small":"gpt-3.5-turbo","agent":"gpt-4o"},"embeddingUrl":"https://api.openai.com/v1/embeddings","embeddings":{"default":"text-embedding-ada-002","large":"text-embedding-3-large","small":"text-embedding-3-small"},"createMessageObject":"getMessageObject","stream":true,"jsonMode":true,"maxTokens":true,"hasEmbeddings":true}
+{
+  "url": "https://api.openai.com/v1/chat/completions",
+  "model": {
+    "default": "gpt-4o-mini",
+    "large": "gpt-4o",
+    "small": "gpt-4o-mini",
+    "agent": "gpt-4o"
+  },
+  "embeddingUrl": "https://api.openai.com/v1/embeddings",
+  "embeddings": {
+    "default": "text-embedding-ada-002",
+    "large": "text-embedding-3-large",
+    "small": "text-embedding-3-small"
+  },
+  "createMessageObject": "getMessageObject",
+  "stream": true,
+  "jsonMode": true,
+  "maxTokens": true,
+  "hasEmbeddings": true
+}

src/config/providers/perplexity.json

@@ -1 +1,14 @@
-{"url":"https://api.perplexity.ai/chat/completions","model":{"default":"llama-3-sonar-large-32k-online","large":"llama-3-sonar-large-32k-online","small":"llama-3-sonar-small-32k-online","agent":"llama-3-sonar-large-32k-online"},"createMessageObject":"getMessageObject","stream":true,"jsonMode":false,"maxTokens":true,"hasEmbeddings":false}
+{
+  "url": "https://api.perplexity.ai/chat/completions",
+  "model": {
+    "default": "llama-3.1-sonar-large-128k-chat",
+    "large": "llama-3.1-sonar-large-128k-chat",
+    "small": "llama-3.1-sonar-small-128k-chat",
+    "agent": "llama-3.1-70b-instruct"
+  },
+  "createMessageObject": "getMessageObject",
+  "stream": true,
+  "jsonMode": false,
+  "maxTokens": true,
+  "hasEmbeddings": false
+}

src/utils/retryWithBackoff.js

@@ -26,9 +26,16 @@ async function retryWithBackoff(fn, options, errorType) {
       let response = await fn();
       if (response?.results) {
         const end = hrtime(start);
+        const resultsEnd = hrtime(start);
+
         const milliseconds = end[0] * 1e3 + end[1] / 1e6;
         response.total_time = milliseconds.toFixed(5);
+
+        const resultsMilliseconds = resultsEnd[0] * 1e3 + resultsEnd[1] / 1e6;
+        response.request_time = resultsMilliseconds.toFixed(5);
+
         response.retries = currentRetry;
+
         return response;
       }
     } catch (error) {