fix(openai): match longest model prefix for context window size

[Kropiunig]

Summary

OpenAICompletion.get_context_window_size() iterates the model-prefix table in insertion order and returns on the first startswith match:

context_windows = {
    "gpt-4": 8192,
    "gpt-4o": 128000,
    "gpt-4o-mini": 200000,
    "gpt-4-turbo": 128000,
    "gpt-4.1": 1047576,
    ...
}
for model_prefix, size in context_windows.items():
    if self.model.startswith(model_prefix):
        return int(size * CONTEXT_WINDOW_USAGE_RATIO)

Because "gpt-4" comes first and "gpt-4o".startswith("gpt-4") is True, every gpt-4* model collapses to gpt-4's 8192-token window before its own, more-specific entry is ever reached. The more specific keys (gpt-4o, gpt-4o-mini, gpt-4-turbo, gpt-4.1, ...) are effectively dead code.

This hits the default model gpt-4o (model: str = "gpt-4o"), so out of the box get_context_window_size() returns int(8192 * 0.85) = 6963 instead of int(128000 * 0.85) = 108800 — making summarization/truncation far too aggressive and causing spurious context-window-exceeded handling.

model	before	after
gpt-4	6963	6963
gpt-4o	6963	108800
gpt-4o-mini	6963	170000
gpt-4-turbo	6963	108800
gpt-4.1	6963	890439

Fix

Sort the prefixes by length (longest first) before matching, so the most specific prefix wins. This mirrors AzureCompletion.get_context_window_size(), which already does exactly this:

for model_prefix, size in sorted(
    context_windows.items(), key=lambda x: len(x[0]), reverse=True
):

Test

Adds test_openai_context_window_uses_longest_prefix_match, asserting gpt-4o -> 128k and gpt-4o-mini -> 200k windows while gpt-4 stays at 8k. It fails on main (6963 == 108800 AssertionError) and passes with this change. Existing behavior for models without a shorter conflicting prefix (e.g. o3-mini -> 200k) is unchanged.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 315f0edd-6524-4e11-b2aa-468f1d9c6c88

📥 Commits

Reviewing files that changed from the base of the PR and between 4c7e483 and f32b5fa.

📒 Files selected for processing (2)

• lib/crewai/src/crewai/llms/providers/openai/completion.py
• lib/crewai/tests/llms/openai/test_openai.py

---

📝 Walkthrough

Walkthrough

OpenAI context-window lookup now prioritizes the longest matching model prefix. A regression test covers gpt-4, gpt-4o, and gpt-4o-mini resolution and expected context-window sizes.

Changes

OpenAI context window matching

Layer / File(s)	Summary
Longest-prefix matching and regression coverage lib/crewai/src/crewai/llms/providers/openai/completion.py, lib/crewai/tests/llms/openai/test_openai.py	get_context_window_size() checks longer prefixes first, while regression tests verify correct sizing for overlapping OpenAI model prefixes.

Suggested reviewers: vinibrsl

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main change: using the longest matching OpenAI model prefix for context window sizing.
Description check	✅ Passed	The description is directly related to the change and accurately explains the bug, fix, and regression test.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}