# Auto-select model heuristics? **URL:** \ **Category:** Discussions\ **Created:** [March 11, 2025, 11:16am UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412 "2025-03-11T11:16:56Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![murnanedaniel](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/murnanedaniel/32/22009_2.png) [@murnanedaniel](https://forum.cursor.com/u/murnanedaniel)\ **Post date:** [March 11, 2025, 11:16am UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412/1 "2025-03-11T11:16:56Z")
The new functionality of creating and editing agent/chat is really powerful, and so is potentially the auto-select model concept. But is there any information from the Cursor team on how the model is actually selected? What are the heuristics? My preference would honestly just be Claude 3.5/3.7, except when there are rate limits or unavailabilities, then fallback to some other model that the user specifies. I think it should be kept simple! ---
**Author:** ![lwc](https://avatars.discourse-cdn.com/v4/letter/l/ee59a6/32.png) [@lwc](https://forum.cursor.com/u/lwc)\ **Post date:** [March 17, 2025, 9:14pm UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412/2 "2025-03-17T21:14:40Z")
Not sure everyone will agree with you to have a simple default preference that might cost double the fast requests. For simple tasks why not choose the agent-less yet 0 fast requests `deekseek-v3` for Edit and Ask Modes, and the 1/3 fast requests `o3-mini` for Agent Mode? Then for more complex tasks, choose by `claude-3.7-sonnet` and as a last resort, `claude-3.7-sonnet-thinking`? P.S. Please consider renaming the title word **heuristics** and perhaps the entire phrase into a more SEO friendly one (e.g. “Please clarify how does Cursor auto-select a model”) so that more people will notice this thread. ---
**Author:** ![SoMaCoSF](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somacosf/32/14963_2.png) [@SoMaCoSF](https://forum.cursor.com/u/SoMaCoSF)\ **Post date:** [March 17, 2025, 9:21pm UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412/3 "2025-03-17T21:21:29Z")
what about a “Logical\_Loop” whereby you spray certain things to Bot – but you can create a || PIPELINE {promptA} → [MODEL\_B] → {promptA\_input} → ETC So push a thing to model A, then pipe output to model B, etc and you have a real workFLOW as an input to a model pipes its output to the input of next model, and it logs its transactional along the way… And you can specifiy in RULES on how to comply to the flow… (Aside from RULES, in the UX – we need a LOGS section that is separate from RULES where LOGGING specificity is declared… (I want LOGGING to be a complete feature. Not an inline… Logging is a trace - so just blurt it out as going along, through the filter of asked. A token tax of sorts, ---
**Author:** ![lwc](https://avatars.discourse-cdn.com/v4/letter/l/ee59a6/32.png) [@lwc](https://forum.cursor.com/u/lwc)\ **Post date:** [March 17, 2025, 9:22pm UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412/4 "2025-03-17T21:22:54Z")
Because you may reach the Fast Requests’ limit rather quickly. ---
**Author:** ![lwc](https://avatars.discourse-cdn.com/v4/letter/l/ee59a6/32.png) [@lwc](https://forum.cursor.com/u/lwc)\ **Post date:** [June 5, 2025, 11:52pm UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412/5 "2025-06-05T23:52:34Z")
> [@Auto mode: please inform us which is the exact model being used](https://forum.cursor.com/t/auto-mode-please-inform-us-which-is-the-exact-model-being-used/96733/24): > > When you use Auto Mode, it intelligently routes to a premium model based on multiple factors, including the model’s availability and any downtime from the model providers. Nevertheless, the Docs give 2 clashing versions: [“It does **not** route based on task type”](https://docs.cursor.com/guides/selecting-models#auto-select) VS [“select the premium model best fit for the **immediate task** ”](https://docs.cursor.com/models#auto). So which is it? Does it only care for uptime or also for the specific task? ---
**Author:** ![techybolek](https://avatars.discourse-cdn.com/v4/letter/t/34f0e0/32.png) [@techybolek](https://forum.cursor.com/u/techybolek)\ **Post date:** [April 30, 2025, 4:27pm UTC](https://forum.cursor.com/t/auto-select-model-heuristics/62412/6 "2025-04-30T16:27:03Z")
I’m wondering when to use auto-select vs selecting a specific model. I have a few scenarios where I prefer specifying a reasoning model like o-3, but in general I’m actually not sure what option is the most likely to give me the biggest bank for the buck. How often do you find yourself using auto-select vs specifying a model? IF you don’t use auto-select, what models do you use and for what?