You can deploy this model and start earning money today!

RuoliuYang/
Monet_v2_stage1rep_merged_qwen25vl7b_8gpu

Name: Monet_v2_stage1rep_merged_qwen25vl7b_8gpu
Availability: InStock

safetensors

qwen2_5_vl

Views

No views yet

Monet no-text ViRL39K GRPO checkpoints

This repo stores intermediate checkpoints from no-text Monet GRPO training on ViRL39K.

Load a checkpoint with:

python

from transformers import AutoProcessor
from RL.monet_models.transformers.monet_modeling_qwen2_5_vl import Qwen2_5_VLForConditionalGeneration

repo_id = "RuoliuYang/Monet_v2_stage1rep_merged_qwen25vl7b_8gpu"
subfolder = "checkpoint-1000"
processor = AutoProcessor.from_pretrained(repo_id, subfolder=subfolder, trust_remote_code=True)
model = Qwen2_5_VLForConditionalGeneration.from_pretrained(repo_id, subfolder=subfolder, trust_remote_code=True)