{"version":"2.1.295","anchor":"beta-header-dropped-on-retry-when-1m-context-is-refused","canonical_anchor":"beta-header-dropped-on-retry-when-1m-context-is-refused","heading":"Requests refused for the 1M-token context are retried without it","tier":"notice","area":"Long Context","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295\/e\/beta-header-dropped-on-retry-when-1m-context-is-refused","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.295","markdown":"### Requests refused for the 1M-token context are retried without it\n\nWhen a request asking for the 1-million-token context window is refused, Claude Code can drop that request and send again\n\n**What**\n\nSome requests ask the API for a 1-million-token context window. The context window is how much conversation the model can take in at once. When such a request is refused, Claude Code can now remove the part asking for the larger window and send the request again.\n\n**Why**\n\nA refusal of the larger context window can now be recovered from by retrying with the normal window, instead of ending as an error.\n\n- Area: Long Context\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}