[Feature Request] 3B Model

### 🚀 The feature, motivation and pitch

I have seen that many users have constrained hardware choices/or are converting pdfs at large scales. Would it be beneficial to train a 3B model (based on Qwen2.5-vl)?

Besides the speedup from switching to a smaller model, we can also further tune for speculative decoding methods such as EAGLE which in turn would make inference of the base model faster.

If that sounds reasonable, I would love to help and contribute to creating this.

### Alternatives

_No response_

### Additional context

(For me specifically, I have to convert ~1M pdfs and I don't have enough compute to do this in a reasonable timeframe)