test: route the cost-calc tests through the local-cost-map fixture

[yuneng-berri]

TLDR

Problem this solves:

• 45 cost-calc tests leave LITELLM_LOCAL_MODEL_COST_MAP set session-wide
• they also leave litellm.model_cost swapped to the local snapshot
• the file's own fixture already did this correctly, unused by them

How it solves it:

• those 45 tests request the existing _local_model_cost_map fixture
• the fixture moves to monkeypatch, so setup failures restore too
• drops a sys.path.insert that resolved against the working directory

User Flow

Before: a proxy admin reports a cost discrepancy, and the maintainer checking it locally reads a stale price

1. The maintainer runs the unit tier in one process, cost-calc tests included
2. A later test in that process reads litellm.model_cost and gets the checked-in snapshot, not the live map
3. Any price quoted from that run reflects the snapshot, so the answer sent back can disagree with what the proxy actually charged

After: the same run reads the live map, so the answer matches the proxy

1. The maintainer runs the same unit tier in one process, cost-calc tests included
2. The cost-calc tests put LITELLM_LOCAL_MODEL_COST_MAP and litellm.model_cost back when each one finishes
3. A later test reads the live map, so a quoted price matches what the proxy charged

Relevant issues

Linear ticket

Pre-Submission checklist

• I have added meaningful tests
• The handful of test files covering my change pass locally, e.g. uv run pytest tests/test_litellm/<your_test_file>.py -v. Leave the suites (make test-unit-*, make test-unit) to CI: it finishes in ~15 minutes where a laptop takes an hour or more
• My PR passes all required CI/CD checks (e.g., lint, schema.d.ts sync check, etc.)
• My PR's scope is as isolated as possible; it only solves 1 specific problem
• I have received a Greptile Confidence Score of at least 4/5 before requesting a maintainer review (Greptile reviews automatically once the PR is opened; only comment @greptileai to re-request a review after pushing changes)

Delays in PR merge?

If you're seeing a delay in your PR being merged, ping the LiteLLM Team on Slack (#pr-review).

Screenshots / Proof of Fix

The leak is only visible from outside the pytest process, so the probe runs the file in-process and reads the environment back afterwards. LITELLM_LOCAL_MODEL_COST_MAP is not part of the conftest snapshot, so nothing else restores it.

# leakprobe.py
import os, pytest
os.environ.pop("LITELLM_LOCAL_MODEL_COST_MAP", None)
pytest.main(["tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py", "-q", "-p", "no:randomly"])
print("LITELLM_LOCAL_MODEL_COST_MAP after:", repr(os.environ.get("LITELLM_LOCAL_MODEL_COST_MAP")))

Before (ff02d5c)

1. python leakprobe.py
2. 174 passed, then LITELLM_LOCAL_MODEL_COST_MAP after: 'True'

After (7c1c800)

1. python leakprobe.py
2. 174 passed, then LITELLM_LOCAL_MODEL_COST_MAP after: None

Type

🧹 Refactoring
✅ Test

Caveats (if any)

• litellm.model_cost restores via the conftest snapshot either way

Final Attestation

• The tests check the right things, including the edge cases, and regressions in the respective real-world customer use-cases are not possible after this PR

[greptile-apps]

Greptile Summary

The PR routes cost-calculation tests through the existing monkeypatch-based local-cost-map fixture so environment and global pricing state are restored after each test.

• Replaces repeated environment and litellm.model_cost assignments with _local_model_cost_map.
• Removes a working-directory-dependent sys.path modification.
• Ratchets the corresponding test-quality budgets downward.

Confidence Score: 5/5

The PR appears safe to merge, with the test-state cleanup correctly scoped and no actionable regressions identified.

The fixture installs a fresh local cost map for each requesting test and relies on pytest monkeypatch teardown to restore both the environment and the original module attribute; the removed path insertion was not load-bearing under supported test invocation.

Important Files Changed

Filename	Overview
tests/test_litellm/litellm_core_utils/llm_cost_calc/test_llm_cost_calc_utils.py	Consolidates local cost-map setup into a monkeypatch fixture that restores environment and module state without weakening the existing assertions.
test-quality-budget.json	Ratchets the path-mutation, environment-mutation, and module-global-mutation budgets to reflect violations removed by the test refactor.

_{Reviews (1): Last reviewed commit: "test: route the cost-calc tests through ..." | Re-trigger Greptile}

[yuneng-berri]

Superseded by #37815, which already covers this file and more

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!