Kostiantyn Sytnyk

@albertvillanova @clefourrier Is there any way to resolve this issue or maybe set the instruction to set the output token limit to ensure it doesn't throw errors and works appropriately even with certain limitations?
PS. Working with limits is MUCH better than not working at all.

I am currently exploring Open-Source alternatives to OpenAI DeepResearch as it's a really great product, one of the most useful since ChatGPT was launched in 2022 for me, as the results of research are incredible high-quality not just simple research with "search" function.
I've decided to try out this Open Deep Research via Hugging face space and ran into issue with the returned output exceeded 128K token limit: