Symbioza runs your containerized GPU job on a rented cloud machine and hands back the files it writes, under a spending limit you set. Describe the job (a container image, the command, links to its input data and a budget) and Symbioza gives you a free estimate first: the expected price, the time limit that applies and whether your prepaid credit covers it. Nothing runs until you approve the submit. Symbioza sizes the machine, runs the job and reports its progress, then returns the output files with their SHA-256 hashes and time-limited download links, and the charge. The budget is a hard ceiling: it is checked before a machine is booked and again before every retry, and you are never billed more than it. Use it for training, fine-tuning, evaluation, batch inference and preprocessing. Credit is added on the Symbioza website.
How to use
- 1Add it in ChatGPT
Open Symbioza in ChatGPT's Plugin directory (button above), select + to add it and sign in if asked.
- 2Use it in a chat
Type @ in the composer and pick Symbioza, or start your message with its name. ChatGPT may also suggest it when it fits what you ask.
Skills 1
- run-on-symbioza
Run a containerized batch job — model training, fine-tuning, evaluation, batch inference, preprocessing — on a rented cloud GPU through the Symbioza connector, under a hard dollar budget. Use when the job needs a GPU this machine lacks or more data or scratch space than local disk holds, or when the user asks to run it remotely with a spending cap. Not for local commands, notebooks, SSH sessions or serving live traffic. Covers what to ask, the free estimate, submitting only after the user approves, polling and collecting files.
Things to ask
- “Estimate what it would cost to fine-tune my model on a cloud GPU with a $20 spending limit.”
- “Run my evaluation harness against this dataset under a $10 budget and bring back the report.”
- “How is my Symbioza job doing, and are the results ready to download?”