{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreihel6hwyvpt4odxp2onpgbc3ygeinljcpkibaoa2gp5f4pwdbdbu4",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mlnvhvjojbe2"
  },
  "path": "/t/batch-api-with-o3-deep-research-spawning-duplicates/1380738#post_1",
  "publishedAt": "2026-05-12T13:19:49.000Z",
  "site": "https://community.openai.com",
  "textContent": "Looking for help understanding what we’re seeing on the Batch API with o3-deep-research-2025-06-26.\n\n**What we submitted**\n\n  * One day of legitimate state-research work across 7 different state slugs\n\n  * Each state: 4 separate POST /v1/batches, each batch containing exactly 1 line targeting /v1/responses\n\n  * Total legitimate request lines submitted across the whole day: 28 (one slug had only 3, so 27)\n\n  * Our server logs the route invocation each time and confirms 28 distinct submissions, no client-side retries\n\n\n\n\n**What we observed in the OpenAI dashboard**\n\nFor the most recently-affected slug alone, the **Logs → Responses** view shows **629 entries** today, all with the JSONL body our 4 batches for that slug submitted (same prompt text, same max_output_tokens: 100000, same\n\nmax_tool_calls: 30, same tool config, background: false, metadata: {}). About **80+ of them are full completions with returned content** (the rest are from 429s after we hit the per-model TPD cap).\n\nSo 4 batch lines became ~80 billed Deep Research completions, plus several hundred additional /v1/responses log entries that didn’t return content. The same pattern played out earlier in the day across the other 6 state\n\nslugs.\n\n**Today’s spend on the model: $47.12** — consistent with dozens of completed Deep Research runs, not 28.\n\n**What we have ruled out from our side**\n\n  * We have exactly two /v1/responses submission paths in our code: a direct background: true call and the batch JSONL above. Every one of the mystery resp_… objects has background: false, so the direct path is not the\n\n\n\nsource.\n\n  * GET /v1/batches?limit=100 returns 27 batches for the day total. No extra batches on the affected slug. No older Wyoming batches. So fan-out is not “duplicate submissions we forgot about”.\n\n  * Server-side logs show exactly one route invocation per slug for the day. No client-side retry loop.\n\n  * Webhooks and crons in the codebase are read-only against OpenAI.\n\n\n\n\n**Other oddities**\n\n  * Each batch’s request_counts reads {total: 1, completed: 0, failed: 0}, despite the dashboard showing many /v1/responses executions associated with that batch’s JSONL body. So the counters don’t reflect what actually ran.\n\n  * output_file_id and error_file_id are both null on all the batches in question, so the work isn’t surfacing through normal batch output channels even though we’re being billed for it.\n\n  * 4 batches for one slug have been stuck in cancelling for 11+ hours after POST /v1/batches/{id}/cancel returned HTTP 200. They never transitioned to cancelled. The other 23 batches show status completed despite having no\n\n\n\n\noutput file.\n\n  * The spawned resp_… objects carry no link back to a parent batch — metadata: {} on every one of them — so there’s no way from a response in the logs to figure out which batch produced it.\n\n\n\n**Questions for the community / OpenAI staff**\n\n1. Under what circumstances does a single batch line with total: 1 produce many /v1/responses executions? Is there an internal retry / fan-out policy on the batch worker?\n\n2. Why do batch request_counts not reflect the actual number of /v1/responses executions performed for that batch?\n\n3. Is it expected that a completed batch with no output_file_id still ran billable executions whose outputs were not delivered through the batch API?\n\n4. Should per-line /v1/responses objects inherit the batch envelope’s metadata so they’re attributable? Right now the response object stands alone with no parent reference, which makes incidents like this very hard to\n\ndiagnose.",
  "title": "Batch API with o3-deep-research spawning duplicates"
}