#Warp repeat itself and duplicate code
1 messages · Page 1 of 1 (latest)
Hi @short ravine. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
I understand your frustration with Warp duplicating code and getting stuck in loops, which is consuming your tokens without providing useful results. This is actually a combination of two separate issues that have been reported by other Warp users.
Issue 1: Command Duplication
The first issue is related to Warp duplicating commands in the terminal. This is a known bug that has been reported in multiple GitHub issues:
Issue #4664 and Issue #4060 describe how commands can appear duplicated in the completed block after execution. For example, running git status might show up as git statgit status in the output.
Issue 2: Token Consumption Despite "Unlimited" Plan
The second issue relates to token limits on the Team plan. While the Team plan does offer unlimited AI requests, there's still a monthly token limit of 70 million tokens. This distinction between requests and tokens is important:
- AI Requests: The number of times you can use Warp AI features (unlimited on Team plan)
- Tokens: The amount of text processed by the AI model (limited to 70M per month on Team plan)
As mentioned in GitHub issue #6300, even Team plan users can hit this token limit.
Why This Is Happening
When Warp AI gets into a loop of detecting duplicated code, creating diffs, and then finding more duplicates, it's consuming tokens with each iteration. This behavior has been reported by other users in Discord support threads, where users noted that the AI "sometimes goes over the same thing several times and corrects its own work because it was wrong the first time then duplicates things in a file which it then needs to go back and fix only to add another duplicate entry."
What You Can Do
- Contact Warp support directly at [email protected] as a paying customer to report this specific issue
- Try breaking down your requests into smaller, more specific tasks
- Be more concise in your prompts to reduce token usage
- Consider restarting Warp when it gets into these loops
The Warp team is aware of these issues and is working on improvements, but reaching out to their support directly is your best option for immediate assistance.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
@short ravine Kapa.ai is still learning and improving, please let me know how I did by reacting below.
This is a duplicate issue, I have already opened issues in github... and have posted in other thread about this exact issue.
I have two task with issues I'm seeing and what you are experiencing is close enough if not exactly what I am:
https://github.com/warpdotdev/Warp/issues/6311
https://github.com/warpdotdev/Warp/issues/6160
Pre-submit Checks I have searched Warp bugs and there are no duplicates I have searched Warp known issues page and my issue is not there I have included the logs (optional, but helps expedite the b...
I've notified the team about these issues, its on our radar and we're looking into fixes. Sometimes its hard to get LLMs to behave in the way that we need exactly so good prompting helps alot. You can also use the rules to help with this potentially. https://docs.warp.dev/features/warp-ai/knowledge
This is insanely frustrating. I hot my 1 month token cap in 10 minutes because of this. More than willing to provide a link to the session. It is severly disappointing that I hit a monthly cap after 10 minutes of use. Further, there is not a clear location that tells me what my current usage is or where the cap is.
Also the default being o1 when o1 is using 4x is also insane.