PRIME-RL/P1-VL-235B-A22B

<div align="center"> <h1 style="font-size: 2em; font-weight: bold;">P1-VL: Bridging Visual Perception and Scientific Reasoning in Physics Olympiads</h1> </div>

<p align="center"> <a href="https://huggingface.co/papers/2602.09443"><b>📜 Paper</b></a> | <a href="https://github.com/PRIME-RL/P1-VL"><b>💻 Code</b></a> | <a href="https://prime-rl.github.io/P1-VL/"><b>🌐 Project Page</b></a> | <a href="https://phyarena.github.io/"><b>🏆 HiPhO Leaderboard</b></a> </p>

<p align="center"> <img src="hipho.png" style="width: 800px" align=center> </p>

<p align="center"> <i>Flagship vision-language model achieving No.3 performance in physics reasoning</i> </p>

Model Description

P1-VL-235B-A22B is the flagship variant of the P1-VL series, a high-performance open-source vision-language model specialized in physics reasoning. It was introduced in P1-VL: Bridging Visual Perception and Scientific Reasoning in Physics Olympiads.

Built on Qwen3-VL-235B-A22B-Thinking and refined through multi-stage reinforcement learning on curated physics competition data, P1-VL-235B-A22B becomes the first open-source Vision-Language Model (VLM) to secure 12 gold medals on HiPhO, ranking No.3 in the model leaderboard. The model effectively solves tasks that require precise diagram-to-logic alignment, demonstrating exceptional performance in physics Olympiad competitions.

Key Highlights

• —🥇 HiPhO Excellence: First open-source VLM to secure 12 gold medals, ranking No.3 globally. When augmented with PhysicsMinions, P1-VL-235B-A22B achieves No.2.
• —🏆 IPhO 2025 Gold-tier Performance: Achieving gold medal performance on International Physics Olympiad
• —📊 FrontierScience-Olympiad: Total score of 64.3/100, outperforming text-only sibling by 2.3 points. When augmented with PhysicsMinions, secures state-of-the-art performance among all evaluated open-source models
• —🎯 STEM Generalization: Consistent improvements over base model across math, and multimodal benchmarks

Performance Benchmarks

HiPhO Results

Evaluated on HiPhO, a rigorous benchmark of 13 exams from 2024–2025, P1-VL-235B-A22B demonstrates top-tier physics reasoning capabilities.

<div align="center">

</div>

FrontierScience-Olympiad Benchmark

P1-VL-235B-A22B achieves significant gains over its base counterpart across all three scientific domains. Remarkably, even on this predominantly text-based benchmark, the multimodal P1-VL-235B-A22B outperforms its text-only sibling (P1-235B-A22B) by a margin of 2.3 points.

<div align="center">

</div>

STEM Benchmarks

Beyond physics reasoning, P1-VL-235B-A22B demonstrates strong generalization across multiple domains, consistently outperforming its base model Qwen3-VL-235B-A22B-Thinking on both text-only and multimodal benchmarks.

<div align="center">

</div>

Usage

from transformers import Qwen3VLMoeForConditionalGeneration, AutoProcessor
from PIL import Image

model_name = "PRIME-RL/P1-VL-235B-A22B"

# Load model and processor
model = Qwen3VLMoeForConditionalGeneration.from_pretrained(
    model_name, dtype="auto", device_map="auto"
)
processor = AutoProcessor.from_pretrained(model_name)

# Load diagram image
image = Image.open("physics_diagram.png")

# Physics problem with visual input
messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": image,
            },
            {
                "type": "text",
                "text": """Analyze this physics diagram and solve the problem:

A block of mass m is placed on an inclined plane with angle θ.
The coefficient of kinetic friction is μ.
Calculate the acceleration of the block down the incline.""",
            },
        ],
    }
]

# Preparation for inference
inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_dict=True,
    return_tensors="pt"
)

# Inference: Generation of the output
generated_ids = model.generate(**inputs, max_new_tokens=8192)
generated_ids_trimmed = [
    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print(output_text[0])

🙏 Acknowledgements

We are grateful to the open-source community for their invaluable contributions. Special thanks to:

• —[Qwen3-VL](https://huggingface.co/collections/Qwen/qwen3-vl) - for providing the foundational base models that powered our research
• —[verl](https://github.com/volcengine/verl) - for the versatile reinforcement learning framework that enabled our training pipeline
• —[vLLM](https://github.com/vllm-project/vllm) - for the efficient LLM serving and inference infrastructure
• —[Megatron-LM](https://github.com/NVIDIA/Megatron-LM) - for the large-scale model training framework

Citation

@misc{p1vl2025,
  title={P1-VL: Bridging Visual Perception and Scientific Reasoning in Physics Olympiads},
  author={P1 Team},
  year={2026},
  url={https://arxiv.org/abs/2602.09443}
}