github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2026-07-01 14:47:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

test(openai): use gpt-4o for image token counting (#38089)

Browse Source 
Image token counting integration coverage is pinned back to `gpt-4o`,
whose usage metadata matches the local vision token estimator. A recent
model refresh moved these checks to `gpt-4.1-mini`, which reports
different live image token usage and broke the exact equality
assertions.
This commit is contained in:
Mason Daugherty
2026-06-11 20:27:18 -04:00
committed by GitHub
parent 76c32db9b3
commit 95d1633341
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
libs/partners/openai/tests/integration_tests/chat_models/test_base.py
View File

@@ -708,7 +708,7 @@ async def test_openai_response_headers_async(use_responses_api: bool) -> None:
def test_image_token_counting_jpeg() -> None:
model = ChatOpenAI(model="gpt-4.1-mini", temperature=0)
model = ChatOpenAI(model="gpt-4o", temperature=0)
image_url = "https://raw.githubusercontent.com/langchain-ai/docs/9f99bb977307a1bd5efeb8dc6b67eb13904c4af1/src/oss/images/checkpoints.jpg"
message = HumanMessage(
content=[
@@ -742,7 +742,7 @@ def test_image_token_counting_jpeg() -> None:
def test_image_token_counting_png() -> None:
model = ChatOpenAI(model="gpt-4.1-mini", temperature=0)
model = ChatOpenAI(model="gpt-4o", temperature=0)
image_url = "https://raw.githubusercontent.com/langchain-ai/docs/4d11d08b6b0e210bd456943f7a22febbd168b543/src/images/agentic-rag-output.png"
message = HumanMessage(
content=[