Answer: Use the --thinking-tokens switch to request a certain number of thinking tokens, specifying the token budget like "1024", "1k", "8k" or "0.01M". Use "0" to disable thinking tokens. Status: Confirmed Entry: tool/aider/models/thinking-tokens-setting Revision: 1 Digest: sha256:02badbc16b3e4c36 (changes whenever the content changes) Timeline: Created: 2026-11-13T02:56:14Z Last changed: 2026-11-13T02:56:14Z Last verified: 2026-11-20T11:42:19Z Next check: 2026-11-21T11:42:19Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Unit | Source | As of switch | --thinking-tokens | | [1] | 2026-11-20 example token budget | 1024 | tokens | [1] | 2026-11-20 disable value | 0 | | [1] | 2026-11-20 useful model | Sonnet 3.7 | | [1] | 2026-11-20 Sources: [1] Aider. "aider.chat/docs/config/reasoning.html". https://aider.chat/docs/config/reasoning.html Role: primary. Accessed: 2026-11-20T11:42:19Z Quote: "You can use the --thinking-tokens switch to request the model use a certain number of thinking tokens." (Supports: switch) Quote: "You can specify the token budget like “1024”, “1k”, “8k” or “0.01M”." (Supports: example token budget) Quote: "Use “0” to disable thinking tokens." (Supports: disable value) Quote: "This switch is useful for Sonnet 3.7." (Supports: useful model) Published: 2026-04-25T16:45:23Z Snapshot SHA-256: d13a58b25ea8a5f41e48af01102f03c2c3c5febb3611db6511235e1b64c438fd (captured 2026-11-13T02:56:13Z) 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 2026-11-13T02:56:14Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T11:42:19Z. No independent second source is configured for this entry Last successful check: 2026-11-20T11:42:19Z Last attempt: 2026-11-20T11:42:19Z Notes and limitations: Extracted automatically from a single official source (section "Thinking tokens"). Related: https://latestn.com/tool/aider/models/unified-diffs-reduce-laziness How do I make GPT-4 Turbo less lazy when editing files with Aider? https://latestn.com/tool/aider/models/xai-grok-model-names Which xAI Grok models can I use with Aider and what are their model names? https://latestn.com/tool/aider/privacy-and-data/analytics-data-collected What data does Aider analytics collect? https://latestn.com/tool/aider/models/select-bedrock-model How do I select an Amazon Bedrock model in Aider? https://latestn.com/tool/aider/models/specify-model-and-api-key How do I specify which model and API key Aider should use? https://latestn.com/tool/aider/models/speed-benchmark-updates What updates does Aider report about its speed benchmarks for GPT-4 Turbo and gpt-3.5-turbo-1106? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T02:56:14Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/aider/models/thinking-tokens-setting/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/models/thinking-tokens-setting POST: https://latestn.com/dispute/tool/aider/models/thinking-tokens-setting with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/aider/models/thinking-tokens-setting?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I configure thinking tokens for reasoning models in Aider? Revision 1, last verified 2026-11-20T11:42:19Z. https://latestn.com/tool/aider/models/thinking-tokens-setting (accessed 2026-11-21T10:00:45Z). @misc{LatestN_tool_aider_models_thinking_tokens_setting_r1, title = {How do I configure thinking tokens for reasoning models in Aider?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/aider/models/thinking-tokens-setting}}, note = {Revision 1; digest sha256:02badbc16b3e4c36; last verified 2026-11-20T11:42:19Z; accessed 2026-11-21T10:00:45Z} } 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/models/thinking-tokens-setting