OpenHands
OpenHands copied to clipboard
All-Hands-AI
Prompt remotely via API
What problem or use case are you trying to solve?
Want to have an external AI Agent feeding it prompts.
Describe the UX of the solution you'd like
/prompt POST "build me XYZ"
Do you have thoughts on the technical implementation?
No.
Describe alternatives you've considered
None.
Additional context
This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
On second thought it probably would not be /prompt POST "build XYZ"
It would be started on one machine as a local server that has port and proxy options as flags. And then it logs out its addresses when started for access like
localhost:1234 192.168.1.1234 345.234.8.1234
And then you can use the CLI or a chat web interface on another machine that is doing post requests to a specified address you fed in as a flag and everything is posting to that address. Just how like OpenAI chat probably works.
The web ui would obviously parse in the UI for clarity although you can turn on verbose and see all the status codes and headers and responses etc. CLI would log verbosely by default.
Because I want all the spare CPU/GPU cycles I can get because I am testing something else.
On Sat, Jun 15, 2024 at 11:02 AM மனோஜ்குமார் பழனிச்சாமி < @.***> wrote:
Why not accessing the front end on the other machine?
— Reply to this email directly, view it on GitHub https://github.com/OpenDevin/OpenDevin/issues/1716#issuecomment-2169849800, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJUO7WFAYLLQUOEIUVJQPDZHRJO7AVCNFSM6AAAAABHR7QPTSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRZHA2DSOBQGA . You are receiving this because you authored the thread.Message ID: @.***>
Are you using local LLM? If so, how about running LLM only in one system?
I do that and many other things. My computer would not be able to handle that addition.
Why is there resistance to this? I just want to offload the compute and try to let the thing code autonomously.
Starting a server that can accept incoming requests is software dev 101.
I’ve made my request clear. If you don’t want to add it I will move on to another project.
On Sat, Jun 15, 2024 at 11:22 AM மனோஜ்குமார் பழனிச்சாமி < @.***> wrote:
Are you using local LLM? If so, how about running LLM only in one system?
— Reply to this email directly, view it on GitHub https://github.com/OpenDevin/OpenDevin/issues/1716#issuecomment-2169889732, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJUO7UYHIV66Z6CDEMYOATZHRL3XAVCNFSM6AAAAABHR7QPTSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNRZHA4DSNZTGI . You are receiving this because you authored the thread.Message ID: @.***>
Hi Jeff. Thanks for keeping this issue up to date. I've removed the stale label.
We may release the cloud version in the future. But it still need more efforts to achieve it.
This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.
I am also highly interested in the API access to the Devin server. Please let me know if the capability is already there. I have asked this in the discussions as well: https://github.com/All-Hands-AI/OpenHands/discussions/3532
This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.
This issue was closed because it has been stalled for over 30 days with no activity.