The stage load SLO was failing at ~95% errors while RPS looked fine. A fixed
load-mock name could reuse a stale deployment without mock_response, so Locust
hit real OpenAI. Always register a unique mock model, preflight one chat, hit
/v1/chat/completions, and attach a status/exception failure breakdown to the
assert so the next red run is diagnosable.
CodSpeed benchmarks the SDK with no IO, so it can't catch regressions that
only appear under real concurrent load through the full proxy stack (auth,
routing, logging, spend, Postgres, Redis). This adds a Locust load test under
tests/e2e/load that drives concurrent POST /chat/completions traffic against a
mock deployment (litellm_params.mock_response), so the measured throughput
reflects proxy overhead rather than a provider's latency, and asserts an
aggregate RPS SLO with a failure-ratio guard. The test is marked load and the
parent conftest sorts load-marked items last so it never perturbs
latency-sensitive suites. Covers reliability.perf.throughput.under_slo.
