## TL;DR
Diffbot's 'plan quota exceeded' means the month's API credits are spent, and article extraction is one of the pricier calls. Check the dashboard for per-call costs, finish the critical articles with the remaining allowance if any, and move bulk extraction to a cheaper path or the next cycle.

## The error
```text
HTTP 429
{"error": "Plan quota exceeded", "credits_used": 10000, "credits_included": 10000}
```

## Fix it
1. Open the Diffbot dashboard and confirm credits_used equals the plan allowance and when the cycle resets.
   Expected: Expected: you know exactly how many credits remain and the reset date.
2. Rank the remaining articles by priority and extract only the critical ones if any credits are left; defer the rest.
   Expected: Expected: the highest-value articles get extracted within the remaining budget.
3. For the deferred bulk, switch to a cheaper extraction path: readability-style extraction locally or a lower-cost provider for simple articles.
   Expected: Expected: the bulk completes without waiting for the reset.
4. Add credit tracking to the pipeline: check the account endpoint before each batch and alert at 80 percent usage.
   Expected: Expected: the next cycle warns before the quota wall instead of hitting it mid-batch.

## When to use
- Your article-extraction batch dies on Diffbot quota mid-run.
- You need to triage which articles are worth the remaining credits.
- You run bulk extraction on a fixed monthly credit pool.

## When not to use
- The error is 401 - that is the token, not quota.
- Every call fails from the start of the cycle - then the plan may not include the Article API.
- You need Diffbot's extraction quality for all articles - then this is an upgrade decision.

## Why it happens
Diffbot bills per API call against a monthly credit pool, and the Article API is a premium call type. Bulk batches consume the pool linearly, and the 429-style quota error is the hard stop. Partial results from earlier in the batch remain valid.

## Variant phrasings
### diffbot plan quota exceeded
Same failure and fix; some searchers phrase the query this way.
### diffbot api credits exhausted article extract
Same failure and fix; some searchers phrase the query this way.

## Edge cases
- Failed extractions (timeouts, blocked pages) still consume credits.
- The dashboard usage can lag real usage by minutes; the API error is the ground truth.

## Provenance

Resolved from the public thread: https://vectle.com/posts/pst_R81xATsbBICXDbBAINcm8w