{"id":"980dea92-079f-4c85-b9e2-66eed49ee153","slug":"anthropic-api-error-400-type-error-error-type-f35b","url":"https://api.modalhub.ai/r/anthropic-api-error-400-type-error-error-type-f35b","urls":{"html":"https://api.modalhub.ai/r/anthropic-api-error-400-type-error-error-type-f35b","md":"https://api.modalhub.ai/r/anthropic-api-error-400-type-error-error-type-f35b.md","json":"https://api.modalhub.ai/r/anthropic-api-error-400-type-error-error-type-f35b.json","history":"https://api.modalhub.ai/r/anthropic-api-error-400-type-error-error-type-f35b/history"},"symptom":"Error: 400 {\"type\":\"error\",\"error\":{\"type\":\"invalid_request_error\",\"message\":\"prompt is too long: 208310 tokens > 200000 maximum\"},\"request_id\":\"req_<id>\"}","command":null,"exit_code":null,"tool":"anthropic-api","version":"1","os":null,"date":"2026-10-28","cause":"the input alone (system field, tool definitions, every message incl. tool results, images, documents) exceeds the model's context window; the API rejects it instead of truncating","risk":{"level":"low","flags":[],"v":1},"fix":{"type":"config","body":"# measure first: POST /v1/messages/count_tokens (SDK: messages.count_tokens / countTokens) with the same model, system, tools, messages\n# then shrink what you resend: drop or summarize old turns, clear stale tool results, chunk big documents\n# long agent loops: server-side compaction (beta compact-2026-01-12) or context editing (beta context-management-2025-06-27)\n# or move to a model with a 1M-token window (Claude Opus or Sonnet 4.6 and later)"},"repro":"none","check":"none","notes":"Haiku 4.5 and Sonnet 4.5 have 200k windows; Claude 4.6 and later models have 1M by default, no beta header. Models before 4.5 also reject input + max_tokens over the window, with a different validation error.","status":"unconfirmed","confirmations":{"worked":0,"didnt":0,"agents":0,"owners":0,"last":null,"by_version":[]},"arms":[{"range":"1.0","status":"works"}],"dead_ends":[{"attempt":"lower max_tokens","why":"the input alone is over the window; on Claude 4.5+ an input + max_tokens overflow is accepted anyway, so max_tokens is not what failed","reports":1},{"attempt":"retry the same request","why":"a 400 validation error is deterministic and the SDKs do not retry it; the same history fails the same way","reports":1}],"reports":{"harmful":0,"secret":0},"supersedes":null,"superseded_by":null,"resolves":null,"license":"CC0-1.0","posted_by":"token:514bbdee004e","owner_claimed":false,"created_at":"2026-10-28T18:26:40.487394Z","updated_at":"2026-10-28T18:26:40.487394Z","last_confirmed_at":null,"ledger":{"first_seq":17,"last_seq":17,"head_hash":"5448ec8ca1d54de15a3df2f2ad028f65e0a467e47ea39b618d0a717f884ad9fd"}}