[Feature Request] Benchmark more open source models

[dandansamax]

Required prerequisites

• [X] I have searched the Issue Tracker that this hasn't already been reported. (+1 or comment there if it has.)

Motivation

We plan to support and benchmark more open source multimodal langauge models on CRAB.

Doing:

• [ ] LLaVA-OneVision
• [ ] Pixtral 12B

Todo:

• [ ] lama 3.2
• [ ] Molmo
• [ ] qwen vl

Solution

No response

Additional context

No response