refactor: update models (#388)

src/client/ernie.rs

@@ -15,7 +15,7 @@ use std::{env, sync::Mutex};
 const API_BASE: &str = "https://aip.baidubce.com/rpc/2.0/ai_custom/v1";
 const ACCESS_TOKEN_URL: &str = "https://aip.baidubce.com/oauth/2.0/token";
 
-const MODELS: [(&str, usize, &str); 6] = [
+const MODELS: [(&str, usize, &str); 7] = [
     // https://cloud.baidu.com/doc/WENXINWORKSHOP/s/clntwmv7t
     ("ernie-4.0-8k", 5120, "/wenxinworkshop/chat/completions_pro"),
     (
@@ -35,6 +35,7 @@ const MODELS: [(&str, usize, &str); 6] = [
         "/wenxinworkshop/chat/ernie-speed-128k",
     ),
     ("ernie-lite-8k", 7168, "/wenxinworkshop/chat/ernie-lite-8k"),
+    ("ernie-tiny-8k", 7168, "/wenxinworkshop/chat/ernie-tiny-8k"),
 ];
 
 lazy_static! {

src/client/mistral.rs

@@ -12,11 +12,11 @@ const API_URL: &str = "https://api.mistral.ai/v1/chat/completions";
 
 const MODELS: [(&str, usize, &str); 5] = [
     // https://docs.mistral.ai/platform/endpoints/
-    ("mistral-large-latest", 32000, "text"),
-    ("mistral-medium-latest", 32000, "text"),
-    ("mistral-small-latest", 32000, "text"),
-    ("open-mixtral-8x7b", 32000, "text"),
     ("open-mistral-7b", 32000, "text"),
+    ("open-mixtral-8x7b", 32000, "text"),
+    ("mistral-small-latest", 32000, "text"),
+    ("mistral-medium-latest", 32000, "text"),
+    ("mistral-large-latest", 32000, "text"),
 ];
 
 

src/client/openai.rs

@@ -13,13 +13,13 @@ use serde_json::{json, Value};
 const API_BASE: &str = "https://api.openai.com/v1";
 
 const MODELS: [(&str, usize, &str); 5] = [
-    // https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo
-    ("gpt-4-turbo-preview", 128000, "text"),
-    ("gpt-4-vision-preview", 128000, "text,vision"),
-    ("gpt-4-1106-preview", 128000, "text"),
     // https://platform.openai.com/docs/models/gpt-3-5-turbo
     ("gpt-3.5-turbo", 16385, "text"),
     ("gpt-3.5-turbo-1106", 16385, "text"),
+    // https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo
+    ("gpt-4-turbo-preview", 128000, "text"),
+    ("gpt-4-1106-preview", 128000, "text"),
+    ("gpt-4-vision-preview", 128000, "text,vision"),
 ];
 
 pub const OPENAI_TOKENS_COUNT_FACTORS: TokensCountFactors = (5, 2);

src/client/qianwen.rs

@@ -28,13 +28,13 @@ const API_URL_VL: &str =
 
 const MODELS: [(&str, usize, &str); 6] = [
     // https://help.aliyun.com/zh/dashscope/developer-reference/api-details
+    ("qwen-turbo", 6000, "text"),
+    ("qwen-plus", 30000, "text"),
     ("qwen-max", 6000, "text"),
     ("qwen-max-longcontext", 28000, "text"),
-    ("qwen-plus", 30000, "text"),
-    ("qwen-turbo", 6000, "text"),
     // https://help.aliyun.com/zh/dashscope/developer-reference/tongyi-qianwen-vl-plus-api
-    ("qwen-vl-max", 0, "text,vision"),
     ("qwen-vl-plus", 0, "text,vision"),
+    ("qwen-vl-max", 0, "text,vision"),
 ];
 
 const TOKENS_COUNT_FACTORS: TokensCountFactors = (4, 14);