Raw data shows OpenAI returned 271 overload errors to a Codex pool in 48 hours
rickasaurus · x · 2026-10-20
Geoffrey Huntley published raw monitoring data from his own proxy pool responding to OpenAI's "Selected model is at capacity" inference issues. The dashboard shows: 271 overload failures in the last 48 hours (HTTP 200 streams ended with an error event), a worst hour of 37 (Oct 20, 09:00 UTC), 7 of 8 accounts in the pool hit, and a median client lag of just 5 ms after the proxy saw the error.
Failures cluster in bursts by hour — busy hours with no failures and quiet hours with many — pointing to OpenAI's capacity, not request volume or a bad account. First-hand accounting of a frontier lab's reliability that paying users care about.
More from Infra
- Inference framework economics: payback period reportedly shrinks from 10 months to 4 — oran_ge · 2026-10-20
- New DeepSeek inference stack released: how will it compare to vLLM, Modal, Together? — finbarrtimbers · 2026-10-20
- COUNT Paper: N-gram Counts on Host Memory Equal 57% More Training Compute — pmttyji · 2026-10-20
- Daytona, building an 'agentic cloud,' raises $60M at $1B valuation — jeremyakahn · 2026-10-20
- ObsessionDB launches realtime data warehouse for agents, claiming 100X faster and 20X cheaper than Snowflake — ycombinator · 2026-10-20
- Polars 2.0 makes streaming the default, speeding queries up to 2.8x with zero code changes — KhuyenTran16 · 2026-10-20