Downloads · 30 days
9
16% of all-time downloads
GaviZhou/qwen3-vl-8b-i2c-hallu-v6-32k
qwen3-vl-8b-i2c-hallu-v6-32k is a image-text-to-text model from GaviZhou. 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 apache-2.0.
This is a full-parameter supervised fine-tune of Qwen/Qwen3-VL-8B-Instruct for visual code-error auditing. Given a ground-truth reference image and candidate plotting code, the model is trained to identify discrepanci…
Downloads · 30 days
9
16% of all-time downloads
All-time downloads
57
Public
Parameters
770K
123 GB on disk
Likes
0
Public
Click a slice to open those files.
.pt105 GB · 86%
From the Hugging Face model README
This is a full-parameter supervised fine-tune of
Qwen/Qwen3-VL-8B-Instruct
for visual code-error auditing. Given a ground-truth reference image and
candidate plotting code, the model is trained to identify discrepancies and
return a strict JSON object containing an errors array.
Use this model for research and development workflows that compare a rendered or target figure with candidate code. The desired response format is:
{
"errors": [
{
"category": "error category",
"description": "description of the discrepancy",
"code_span": {
"start_line": 12,
"end_line": 14,
"anchor": "relevant code fragment"
}
}
]
}
code_span is optional. Line numbers are 1-based and should refer to the
candidate code supplied in the prompt.
Qwen/Qwen3-VL-8B-Instructtuner_type=full)1e-5, weight decay 0.01loss_scale=last_round)The checkpoint published here is the best checkpoint from the first training
run: checkpoint-198.
Teacher-forced validation metrics at checkpoint-198:
| Metric | Value |
|---|---|
eval_loss | 0.352678 |
eval_token_acc | 0.900092 |
The packed validation set contains 42 sequences. These are likelihood and token-level metrics, not direct measurements of error-detection quality or hallucination rate. For model selection or production use, additionally evaluate generated answers on held-out examples with JSON/schema validity, error-category precision/recall/F1, and code-anchor/line-alignment metrics.
Install a recent Transformers release with Qwen3-VL support, then load the model as an image-text-to-text model:
import torch
from transformers import AutoProcessor, Qwen3VLForConditionalGeneration
model_id = "GaviZhou/qwen3-vl-8b-i2c-hallu-v6-32k"
model = Qwen3VLForConditionalGeneration.from_pretrained(
model_id,
torch_dtype=torch.bfloat16,
device_map="auto",
)
processor = AutoProcessor.from_pretrained(model_id)
candidate_code = """# Candidate plotting code; retain 1-based line references."""
messages = [
{
"role": "user",
"content": [
{"type": "image", "image": "/path/to/ground_truth.png"},
{
"type": "text",
"text": (
"Compare the ground-truth image with the candidate code. "
"Return only the required JSON object.\n\n"
f"Candidate code:\n{candidate_code}"
),
},
],
}
]
inputs = processor.apply_chat_template(
messages,
tokenize=True,
add_generation_prompt=True,
return_dict=True,
return_tensors="pt",
).to(model.device)
generated_ids = model.generate(**inputs, max_new_tokens=1024)
generated_ids = [
output_ids[len(input_ids):]
for input_ids, output_ids in zip(inputs.input_ids, generated_ids)
]
print(processor.batch_decode(generated_ids, skip_special_tokens=True)[0])
The upstream Qwen3-VL model card contains additional examples and deployment guidance.
This checkpoint is derived from Qwen/Qwen3-VL-8B-Instruct, which is released
under the Apache-2.0 license. Users must also ensure that their use of the
fine-tuning data and downstream inputs is permitted.
If you use this model, please also cite the base model:
@misc{qwen3technicalreport,
title={Qwen3 Technical Report},
author={Qwen Team},
year={2025},
eprint={2505.09388},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2505.09388}
}