hchoi256/mba-business-ideation-agent

MBA: Multimodal Business Ideation Agent

This Space demonstrates MBA (Multimodal Benchmark and Agents for Real-World Business Ideation), a vision-language model fine-tuned from Qwen2-VL-7B to generate structured, market-grounded B2B startup ideas from visual inputs.

How it works

1. 1.Upload an image (texture, scene, UI layout, PCB, etc.)
2. 2.Select the image domain and a business analysis lens
3. 3.The MBA model generates a structured business idea as JSON with:
4. 4.Title: concise business idea name
5. 5.Description: core function, target user, need, and benefit
6. 6.Implementation: how to build or deploy it
7. 7.Differentiation: what makes it defensible

Model variants

• —MBA-k-7B (GRPO): Best overall, tuned for open-ended ideation
• —MBA-b-7B (GRPO): Tuned for closed-branch ideation
• —MBA-7B (SFT): SFT-only baseline
• —Qwen2-VL-7B (Base): Original model without MBA fine-tuning

Source

• —📄 Paper: MBA: Multimodal Benchmark and Agents for Real-World Business Ideation
• —🤗 Model: hchoi256/mba
• —💻 Code: github.com/hchoi256/MBA
• —📊 Dataset: hchoi256/MBA-Bench (Apache-2.0)

Citation

@misc{choi2026mbamultimodalbenchmarkagents,
      title={MBA: Multimodal Benchmark and Agents for Real-World Business Ideation},
      author={Hojun Choi and Jaeyo Shin and Suin Lee and Hyunjung Shim},
      year={2026},
      eprint={2608.11616},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2608.11616},
}