refactor: reorder models (#372)

The capable ones come first.

6 files changed, 16 insertions, 18 deletions

diff --git a/src/client/claude.rs b/src/client/claude.rs
index d1b39d7..1eb83d1 100644
--- a/src/client/claude.rs
+++ b/src/client/claude.rs
@@ -19,14 +19,11 @@ use serde_json::{json, Value};
 
 const API_BASE: &str = "https://api.anthropic.com/v1/messages";
 
-const MODELS: [(&str, usize, &str); 6] = [
+const MODELS: [(&str, usize, &str); 3] = [
     // https://docs.anthropic.com/claude/docs/models-overview
     ("claude-3-opus-20240229", 200000, "text,vision"),
     ("claude-3-sonnet-20240229", 200000, "text,vision"),
     ("claude-3-haiku-20240307", 200000, "text,vision"),
-    ("claude-2.1", 200000, "text"),
-    ("claude-2.0", 100000, "text"),
-    ("claude-instant-1.2", 100000, "text"),
 ];
 
 const TOKENS_COUNT_FACTORS: TokensCountFactors = (5, 2);
diff --git a/src/client/gemini.rs b/src/client/gemini.rs
index 68db15a..60c1e3b 100644
--- a/src/client/gemini.rs
+++ b/src/client/gemini.rs
@@ -14,6 +14,7 @@ const MODELS: [(&str, usize, &str); 2] = [
     // https://ai.google.dev/models/gemini
     ("gemini-pro", 30720, "text"),
     ("gemini-pro-vision", 12288, "vision"),
+    // ("gemini-1.5-pro", 1048576, "text,vision"),
 ];
 
 const TOKENS_COUNT_FACTORS: TokensCountFactors = (5, 2);
diff --git a/src/client/mistral.rs b/src/client/mistral.rs
index 25df830..36478b8 100644
--- a/src/client/mistral.rs
+++ b/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-small-latest", 32000, "text"),
-    ("mistral-medium-latest", 32000, "text"),
     ("mistral-large-latest", 32000, "text"),
-    ("open-mistral-7b", 32000, "text"),
+    ("mistral-medium-latest", 32000, "text"),
+    ("mistral-small-latest", 32000, "text"),
     ("open-mixtral-8x7b", 32000, "text"),
+    ("open-mistral-7b", 32000, "text"),
 ];
 
 
diff --git a/src/client/openai.rs b/src/client/openai.rs
index 1abdd48..a1bd3db 100644
--- a/src/client/openai.rs
+++ b/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-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-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"),
 ];
 
 pub const OPENAI_TOKENS_COUNT_FACTORS: TokensCountFactors = (5, 2);
diff --git a/src/client/qianwen.rs b/src/client/qianwen.rs
index 3e69dae..a95556b 100644
--- a/src/client/qianwen.rs
+++ b/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-plus", 0, "text,vision"),
     ("qwen-vl-max", 0, "text,vision"),
+    ("qwen-vl-plus", 0, "text,vision"),
 ];
 
 const TOKENS_COUNT_FACTORS: TokensCountFactors = (4, 14);
diff --git a/src/client/vertexai.rs b/src/client/vertexai.rs
index ffe9151..ba43796 100644
--- a/src/client/vertexai.rs
+++ b/src/client/vertexai.rs
@@ -14,13 +14,13 @@ use serde::Deserialize;
 use serde_json::{json, Value};
 use std::path::PathBuf;
 
-// https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models
-const MODELS: [(&str, usize, &str); 5] = [
+const MODELS: [(&str, usize, &str); 2] = [
+    // https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models
     ("gemini-1.0-pro", 24568, "text"),
     ("gemini-1.0-pro-vision", 14336, "text,vision"),
-    ("gemini-1.0-ultra", 8192, "text"),
-    ("gemini-1.0-ultra-vision", 8192, "text,vision"),
-    ("gemini-1.5-pro", 1000000, "text"),
+    // ("gemini-1.0-ultra", 8192, "text"),
+    // ("gemini-1.0-ultra-vision", 8192, "text,vision"),
+    // ("gemini-1.5-pro", 1000000, "text"),
 ];
 
 const TOKENS_COUNT_FACTORS: TokensCountFactors = (5, 2);