Merge pull request #22 from pamelafox/imagedetail

Assume auto for detail

Author: pamelafox

.pre-commit-config.yaml

@@ -1,15 +1,15 @@
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.6.0
+    rev: v5.0.0
     hooks:
     -   id: check-yaml
     -   id: end-of-file-fixer
     -   id: trailing-whitespace
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.4.1
+    rev: v0.9.0
     hooks:
     -   id: ruff
 -   repo: https://github.com/psf/black
-    rev: 24.4.0
+    rev: 24.10.0
     hooks:
     -   id: black

CHANGELOG.md

@@ -2,6 +2,10 @@
 
 All notable changes to this project will be documented in this file.
 
+## [0.1.11] - Jan 10, 2025
+
+- When no "detail" is provided for an "image_url" message part, "auto" is now assumed.
+
 ## [0.1.10] - Aug 7, 2024
 
 - Add additional OpenAI.com model names to the `get_token_limit` function.

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "openai-messages-token-helper"
 description = "A helper library for estimating tokens used by messages sent through OpenAI Chat Completions API."
-version = "0.1.10"
+version = "0.1.11"
 authors = [{name = "Pamela Fox"}]
 requires-python = ">=3.9"
 readme = "README.md"

src/openai_messages_token_helper/message_builder.py

@@ -26,7 +26,7 @@ def normalize_content(content: Union[str, Iterable[ChatCompletionContentPartPara
         return unicodedata.normalize("NFC", content)
     else:
         for part in content:
-            if "image_url" not in part:
+            if part["type"] == "text":
                 part["text"] = unicodedata.normalize("NFC", part["text"])
         return content
 

src/openai_messages_token_helper/model_helper.py

@@ -115,7 +115,9 @@ def count_tokens_for_message(model: str, message: ChatCompletionMessageParam, de
                 if item["type"] == "text":
                     num_tokens += len(encoding.encode(item["text"]))
                 elif item["type"] == "image_url":
-                    num_tokens += count_tokens_for_image(item["image_url"]["url"], item["image_url"]["detail"], model)
+                    num_tokens += count_tokens_for_image(
+                        item["image_url"]["url"], item["image_url"].get("detail", "auto"), model
+                    )
         elif isinstance(value, str):
             num_tokens += len(encoding.encode(value))
         else:

tests/image_messages.py

@@ -16,6 +16,23 @@
     "count_4o_mini": 8511,
 }
 
+text_and_tiny_image_message_nodetail = {
+    "message": {
+        "role": "user",
+        "content": [
+            {"type": "text", "text": "Describe this picture:"},
+            {
+                "type": "image_url",
+                "image_url": {
+                    "url": "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8z/C/HgAGgwJ/lK3Q6wAAAABJRU5ErkJggg=="
+                },
+            },
+        ],
+    },
+    "count": 266,
+    "count_4o_mini": 8511,
+}
+
 text_and_tiny_image_message_low = {
     "message": {
         "role": "user",
@@ -52,4 +69,9 @@
     "count_4o_mini": 19842,
 }
 
-IMAGE_MESSAGE_COUNTS = [text_and_tiny_image_message, text_and_tiny_image_message_low, text_and_large_image_message]
+IMAGE_MESSAGE_COUNTS = [
+    text_and_tiny_image_message,
+    text_and_tiny_image_message_nodetail,
+    text_and_tiny_image_message_low,
+    text_and_large_image_message,
+]

tests/test_imageshelper.py

@@ -1,6 +1,7 @@
 import base64
 
 import pytest
+
 from openai_messages_token_helper import count_tokens_for_image
 
 

tests/test_messagebuilder.py

@@ -6,6 +6,7 @@
     ChatCompletionToolChoiceOptionParam,
     ChatCompletionToolParam,
 )
+
 from openai_messages_token_helper import build_messages, count_tokens_for_message
 
 from .functions import search_sources_toolchoice_auto

tests/test_modelhelper.py

@@ -1,4 +1,5 @@
 import pytest
+
 from openai_messages_token_helper import count_tokens_for_message, count_tokens_for_system_and_tools, get_token_limit
 
 from .functions import FUNCTION_COUNTS, search_sources_toolchoice_auto