panademo

Would you be able to open-source the script that was used to achieve the 33.6% on SWE-Bench lite using SuperCoder2.0?

K2 Thinking Browsecomp/HLE Reproducibility | 结果复现

1

Hi Moonshot team, thank you so much for open-sourcing such impressive models and sharing your research! Just a question regarding reproducibility of the [K2 Thinking benchmarks](https://moonshotai.github.io/Kimi-K2/thinking.html) with tool-use: How can...

ParallelMuse或Heavy版本有计划公开源码不 / Plans to open source ParallelMuse or Heavy variant?

2

Hey Tongyi team, thank you so much for open-sourcing such impressive models and sharing your research! Are there plans to open source ParallelMuse, or the "Heavy" inference variant? --- 嗨，通义团队，非常感谢你们开源如此令人印象深刻的模型并分享研究成果！请问是否有计划开源...

IterResearch/"Heavy" mode source code? | IterResearch/“Heavy” 模式源码？

3

Hi, thank you so much for open-sourcing such an impressive system🙏! Just wondering where the code for the "IterResearch-based 'Heavy' mode, which uses a test-time scaling strategy to unlock the...

Humanity's Last Exam?

1

Thank you for such a great project/tool and for open-sourcing it. The GAIA and FRAMES scores are very impressive! Are there any plans to also benchmark it on [Humanity's Last...