Answer: Aider works around output token limits using models that support "prefilling": when it hits the output token limit while collecting code edits, it initiates another LLM request with the partial response prefilled so the model continues where it left off, and this can be repeated for very long outputs. Status: Confirmed Entry: tool/aider/features/infinite-output-overview Revision: 1 Digest: sha256:158a523165cbe61d (changes whenever the content changes) Timeline: Created: 2026-11-13T02:56:35Z Last changed: 2026-11-13T02:56:35Z Last verified: 2026-11-20T11:43:30Z Next check: 2026-11-21T11:43:30Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of mechanism | prefilling | [1] | 2026-11-20 action at limit | aider simply initiates another LLM request with the partial response prefilled | [1] | 2026-11-20 repeatability | repeated | [1] | 2026-11-20 feature name | infinite output | [1] | 2026-11-20 Sources: [1] Aider. "aider.chat/docs/more/infinite-output.html". https://aider.chat/docs/more/infinite-output.html Role: primary. Accessed: 2026-11-20T11:43:30Z Quote: "Aider is able to work around this limit with models that support “prefilling” the assistant response." (Supports: mechanism) Quote: "it hits the output token limit, aider simply initiates another LLM request with the partial response prefilled." (Supports: action at limit) Quote: "This prefilling of the partially completed response can be repeated, allowing for very long outputs." (Supports: repeatability) Quote: "Aider supports “infinite output” for models that support “prefill”" (Supports: feature name) Published: 2026-04-25T16:45:23Z Snapshot SHA-256: 4cc7341fc7493a3ed46172cd70fcbe96ac537f098bf6e3ef3e2278209818d9f4 (captured 2026-11-13T02:56:35Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-13T02:56:35Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T11:43:30Z. No independent second source is configured for this entry Last successful check: 2026-11-20T11:43:30Z Last attempt: 2026-11-20T11:43:30Z Notes and limitations: Extracted automatically from a single official source (section "Infinite output"). Related: https://latestn.com/tool/aider/features/make-code-changes How do I make changes to files with Aider? https://latestn.com/tool/aider/features/os-specific-notifications What OS-specific notification options does Aider support? https://latestn.com/tool/aider/features/paste-llm-reply-back How do I paste an LLM's reply back into Aider so it edits my files? https://latestn.com/tool/aider/features/edit-ssh-config How do I use Aider to edit SSH configuration files? https://latestn.com/tool/aider/features/editor-edit-formats What are Aider's editor-diff and editor-whole edit formats used for? https://latestn.com/tool/aider/features/enable-voice-coding How do I enable voice coding in Aider? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T02:56:35Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/aider/features/infinite-output-overview/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/aider/features/infinite-output-overview POST: https://latestn.com/dispute/tool/aider/features/infinite-output-overview with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/aider/features/infinite-output-overview?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How does Aider handle models with output token limits? Revision 1, last verified 2026-11-20T11:43:30Z. https://latestn.com/tool/aider/features/infinite-output-overview (accessed 2026-11-21T09:57:53Z). @misc{LatestN_tool_aider_features_infinite_output_overview_r1, title = {How does Aider handle models with output token limits?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/aider/features/infinite-output-overview}}, note = {Revision 1; digest sha256:158a523165cbe61d; last verified 2026-11-20T11:43:30Z; accessed 2026-11-21T09:57:53Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/aider/features/infinite-output-overview