Would it be possible to consider a UI update in Warp to display the multiplier cost for each LLM? Currently, the Usage section only shows the remaining requests, which feels a bit lacking in transparency. It doesn’t indicate how much each request actually costs, making it difficult to track usage accurately.
#Show LLM multiplier
pure folio
wispy grottoBOT
Hi @pure folio. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Currently, Warp does not display the specific "multiplier cost" or per-model request cost for each LLM in the UI. Multiple users have requested more transparency around how many requests different models consume, and there are several open and closed feature requests asking for a clearer breakdown of AI credit usage, including per-model costs and session summaries. However, Warp team members have stated that it is difficult to quantify exactly how many requests each model will consume, as it depends on the prompt, environment, and task complexity. Generally, more premium models (like Opus 4.1) consume more requests, but there is no published or UI-exposed multiplier for each model at this time. Users are encouraged to empirically test different models to see their relative consumption, but a direct multiplier or cost-per-model is not currently available in the UI or documentation. The Warp team is aware of this feedback and has acknowledged the need for more transparency in this area, but no such feature has been implemented yet Features: hooks, slash commands and maybe .wrap + warp.md, credit costs per session display, and mcp.json.
If you would like to see this feature, you can upvote or comment on the relevant GitHub issues to help prioritize it for future development.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
jaunty holly
@pure folio our update today will show more details on credit (previously called requests) usage per AI interaction for the most recent interaction.
This should help with transparency, but note we dont show the multiplier per model at this stage.