OpenAI Platform · Request logs
OpenAI put frontier models behind one clean API. This rebuild keeps the log's exact structure — and makes a 429 at 9:41 impossible to miss.
| Time | Request | Model | Tokens in | Tokens out | Latency | Status |
|---|---|---|---|---|---|---|
| 09:42:18 | req_9f2c81ab04d1 /v1/chat/completions | gpt-5 | 2,814 | 912 | 3.42s | 200 OK |
| 09:42:11 | req_9f2c7d55e2b8 /v1/chat/completions | gpt-5-mini | 640 | 188 | 780ms | 200 OK |
| 09:41:57 | req_9f2c76c9aa03 /v1/embeddings | text-embedding-3-large | 5,120 | — | 240ms | 200 OK |
| 09:41:44 | req_9f2c71124f77 /v1/chat/completions | gpt-5 | 3,902 | — | 60ms | 429 Rate limit |
| 09:41:31 | req_9f2c6b8d31c5 /v1/chat/completions | o3-mini | 1,204 | 2,048 | 5.91s | 200 OK |
| 09:41:02 | req_9f2c64f0b9e2 /v1/chat/completions | gpt-5-mini | 512 | 96 | 610ms | 200 OK |
| 09:40:48 | req_9f2c5fa7d044 /v1/responses | gpt-5 | 8,110 | 1,744 | 7.28s | 200 OK |
| 09:40:20 | req_9f2c58c2e6a9 /v1/chat/completions | gpt-4.1 | 990 | — | 130ms | 400 Error |
| 09:39:55 | req_9f2c52091bd7 /v1/chat/completions | gpt-5 | 2,201 | — | 45ms | 500 Error |
Rebuilt by Kyre — same layout, same data, same information architecture. Only the execution changed.
What changed — and why it matters
A rate limit isn't a bug — it's capacity you're paying for and not getting. I split it from real failures: amber, not red. An engineer logs both as 4xx/5xx; knowing they're triaged differently is the judgment.
This log answers two questions: what failed, and what's expensive. Token counts lead the row because the cost question comes back every week — the request ID is context, not the story. That's a call about which number gets read first.
Nine visible rows can hide 18,000 requests of trouble. The footer totals failures and rate limits across the whole window — so a clean first page never lies to you. Deciding what the page must admit is the design.