Downloads · 30 days
7
13% of all-time downloads
lzy-vlm-lab-opd/qwen3vl_grpo_mmf_sft6939_lr5e6_steps50
qwen3vl_grpo_mmf_sft6939_lr5e6_steps50 is a image-text-to-text model from lzy-vlm-lab-opd. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as other.
Vision-language model fine-tuned in the OPD / lzy-vlm-lab group from Qwen3-VL-8B (8.77B params).
Downloads · 30 days
7
13% of all-time downloads
All-time downloads
52
Public
Parameters
8.8B
17.5 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors17.5 GB · 100%
From the Hugging Face model README
Vision-language model fine-tuned in the OPD / lzy-vlm-lab group from Qwen3-VL-8B (8.77B params).
bfloat16Qwen3VLForConditionalGeneration (qwen3_vl)GRPO RL initialized from the sft6939 SFT weights, learning rate 5e-6, 50 steps.
model.safetensors — merged single-file inference weightsconfig.json, generation_config.json, tokenizer.json, tokenizer_config.json, processor_config.json, chat_template.jinja — full HF runtime configfrom transformers import AutoModelForVision2Seq, AutoProcessor
model = AutoModelForVision2Seq.from_pretrained("lzy-vlm-lab-opd/qwen3vl_grpo_mmf_sft6939_lr5e6_steps50")
processor = AutoProcessor.from_pretrained("lzy-vlm-lab-opd/qwen3vl_grpo_mmf_sft6939_lr5e6_steps50")
Private repo — requires hf auth login with read access to the org.