From a2cde3051558d3f0c5cb4c3395ecffe00435c4c7 Mon Sep 17 00:00:00 2001
From: J S <49557684+svilupp@users.noreply.github.com>
Date: Wed, 14 Aug 2024 08:54:08 +0100
Subject: [PATCH] Add ChatGPT 4o (#195)

---
 CHANGELOG.md            |  5 +++++
 Project.toml            |  2 +-
 src/llm_openai.jl       | 45 +++++++++++++++++++++++++++++++++++++++++
 src/user_preferences.jl |  6 ++++++
 4 files changed, 57 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 00fc0a244..433a9357e 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -10,6 +10,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Fixed
 
+## [0.49.0]
+
+### Added
+- Added new OpenAI's model "chatgpt-4o-latest" to the model registry with alias "chatgpt". This model represents the latest version of ChatGPT-4o tuned specifically for ChatGPT.
+
 ## [0.48.0]
 
 ### Added
diff --git a/Project.toml b/Project.toml
index c2ca2ab49..d199ad268 100644
--- a/Project.toml
+++ b/Project.toml
@@ -1,7 +1,7 @@
 name = "PromptingTools"
 uuid = "670122d1-24a8-4d70-bfce-740807c42192"
 authors = ["J S @svilupp and contributors"]
-version = "0.48.0"
+version = "0.49.0"
 
 [deps]
 AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"
diff --git a/src/llm_openai.jl b/src/llm_openai.jl
index fe866f022..2f3f51f31 100644
--- a/src/llm_openai.jl
+++ b/src/llm_openai.jl
@@ -683,6 +683,51 @@ const OPENAI_TOKEN_IDS = Dict("true" => 837,
     "39" => 2137,
     "40" => 1272
 )
+# GPT-4o token IDs as per tiktoken
+# "true": 3309,
+# "false": 7556,
+# "unknown": 33936,
+# "other": 2141,
+# "1": 16,
+# "2": 17,
+# "3": 18,
+# "4": 19,
+# "5": 20,
+# "6": 21,
+# "7": 22,
+# "8": 23,
+# "9": 24,
+# "10": 702,
+# "11": 994,
+# "12": 899,
+# "13": 1311,
+# "14": 1265,
+# "15": 1055,
+# "16": 1125,
+# "17": 1422,
+# "18": 1157,
+# "19": 858,
+# "20": 455,
+# "21": 2040,
+# "22": 1709,
+# "23": 1860,
+# "24": 1494,
+# "25": 1161,
+# "26": 2109,
+# "27": 2092,
+# "28": 2029,
+# "29": 2270,
+# "30": 1130,
+# "31": 2911,
+# "32": 1398,
+# "33": 2546,
+# "34": 3020,
+# "35": 2467,
+# "36": 2636,
+# "37": 2991,
+# "38": 3150,
+# "39": 3255,
+# "40": 1723,
 
 """
     encode_choices(schema::OpenAISchema, choices::AbstractVector{<:AbstractString}; kwargs...)
diff --git a/src/user_preferences.jl b/src/user_preferences.jl
index 6f770c88e..935fe1b95 100644
--- a/src/user_preferences.jl
+++ b/src/user_preferences.jl
@@ -330,6 +330,7 @@ aliases = merge(
         "gpt4v" => "gpt-4-vision-preview", # 4v is for "4 vision"
         "gpt4t" => "gpt-4-turbo", # 4t is for "4 turbo"
         "gpt3t" => "gpt-3.5-turbo-0125", # 3t is for "3 turbo"
+        "chatgpt" => "chatgpt-4o-latest",
         "ada" => "text-embedding-ada-002",
         "emb3small" => "text-embedding-3-small",
         "emb3large" => "text-embedding-3-large",
@@ -467,6 +468,11 @@ registry = Dict{String, ModelSpec}(
         1.5e-7,
         6e-7,
         "GPT-4 Omni Mini, the smallest and fastest model based on GPT4 (and cheaper than GPT3.5Turbo). Context of 128K, knowledge until October 2023. Currently points to version gpt-4o-2024-07-18."),
+    "chatgpt-4o-latest" => ModelSpec("chatgpt-4o-latest",
+        OpenAISchema(),
+        5e-6,
+        1.5e-5,
+        "ChatGPT-4o-latest is the latest version of ChatGPT-4o tuned for ChatGPT. It is the NOT same as gpt-4o-latest."),
     "gpt-4-vision-preview" => ModelSpec(
         "gpt-4-vision-preview",
         OpenAISchema(),