[Feature] Support training Qwen2.5-VL-32B eagle model

[gerayking]

Checklist

• 1. If the issue you raised is not a feature but a question, please raise a discussion at https://github.com/sgl-project/SpecForge/discussions/new/choose Otherwise, it will be closed.
• 2. Please use English, otherwise it will be closed.

Motivation

I noticed SpecForge currently supports Qwen2.5-VL-7B via Transformers. I think we should instead add systematic VL support through sglang, rather than a model-specific Transformers integration.

        if (
            args.is_vlm
            and draft_model_config.target_model_type == "qwen2_5_vl"
            and args.tp_size == 1
        ):
            from transformers import Qwen2_5_VLForConditionalGeneration

            target_model = (
                Qwen2_5_VLForConditionalGeneration.from_pretrained(
                    pretrained_model_name_or_path=args.target_model_path,
                    torch_dtype=torch.bfloat16,
                )
                .eval()
                .cuda()
            )

Related resources

No response

[gerayking]

I’m very interested in this—could you please assign it to me?

[jiapingW]

Sure.