# PyVision-RL: Forging Open Agentic Vision Models via RL

arXiv: https://arxiv.org/abs/2602.20739 · HTML: https://mlsanity.com/paper/2602.20739/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal and embodied agents, one of 16 directions in LLM agents and tool use.
- Builds on 53 of its 64 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Placed in

- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 30 of 53)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://mlsanity.com/paper/2203.10244/) (2022)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [Understanding R1-Zero-Like Training: A Critical Perspective](https://mlsanity.com/paper/2503.20783/) (2025)
- [Group Sequence Policy Optimization](https://mlsanity.com/paper/2507.18071/) (2025)
- [Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces](https://mlsanity.com/paper/2412.14171/) (2024)
- [ViperGPT: Visual Inference via Python Execution for Reasoning](https://mlsanity.com/paper/2303.08128/) (2023)
- [MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?](https://mlsanity.com/paper/2403.14624/) (2024)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset](https://mlsanity.com/paper/2402.14804/) (2024)
- [MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action](https://mlsanity.com/paper/2303.11381/) (2023)
- [Visual Programming: Compositional visual reasoning without training](https://mlsanity.com/paper/2211.11559/) (2022)
- [Video-R1: Reinforcing Video Reasoning in MLLMs](https://mlsanity.com/paper/2503.21776/) (2025)
- [V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs](https://mlsanity.com/paper/2312.14135/) (2023)
- [InfographicVQA](https://mlsanity.com/paper/2104.12756/) (2021)
- [VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning](https://mlsanity.com/paper/2504.08837/) (2025)
- [LongVILA: Scaling Long-Context Visual Language Models for Long Videos](https://mlsanity.com/paper/2408.10188/) (2024)
- [MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning](https://mlsanity.com/paper/2503.07365/) (2025)
- [We-Math: Does Your Large Multimodal Model Achieve Human-like Mathematical Reasoning?](https://mlsanity.com/paper/2407.01284/) (2024)
- [Pixel Reasoner: Incentivizing Pixel-Space Reasoning with Curiosity-Driven Reinforcement Learning](https://mlsanity.com/paper/2505.15966/) (2025)
- [MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention](https://mlsanity.com/paper/2506.13585/) (2025)
- [Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization](https://mlsanity.com/paper/2411.10442/) (2024)
- [MiMo-V2-Flash Technical Report](https://mlsanity.com/paper/2601.02780/) (2026)
- [Thyme: Think Beyond Images](https://mlsanity.com/paper/2508.11630/) (2025)
- [Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models](https://mlsanity.com/paper/2408.15556/) (2024)
- [SpaceR: Reinforcing MLLMs in Video Spatial Reasoning](https://mlsanity.com/paper/2504.01805/) (2025)
- [DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models](https://mlsanity.com/paper/2411.00836/) (2024)
- [SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning](https://mlsanity.com/paper/2509.02479/) (2025)

## Built on by (most cited first, 16 of 16)

- [TACO: Tool-Augmented Credit Optimization for Agentic Tool Use](https://mlsanity.com/paper/2606.30251/) (2026)
- [EVE: Verifiable Self-Evolution of MLLMs via Executable Visual Transformations](https://mlsanity.com/paper/2604.18320/) (2026)
- [Agent Explorative Policy Optimization for Multimodal Agentic Reasoning](https://mlsanity.com/paper/2605.28774/) (2026)
- [Perceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning](https://mlsanity.com/paper/2607.01191/) (2026)
- [Self-Evolving Code-with-Image Reasoning](https://mlsanity.com/paper/2608.11292/) (2026)
- [Routing Before Looking: Query-Adaptive Evidence Acquisition for Long-form Video Understanding](https://mlsanity.com/paper/2608.20805/) (2026)
- [SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning](https://mlsanity.com/paper/2603.23483/) (2026)
- [OPD-Evolver: Cultivating Holistic Agent Evolver via On-Policy Distillation](https://mlsanity.com/paper/2606.17628/) (2026)
- [FaithEyes: Towards Faithful Tool Use via Multi-Agent Process-Image Verification](https://mlsanity.com/paper/2607.28225/) (2026)
- [Beacon: Knowing When and How to Perform Agentic Visual Reasoning](https://mlsanity.com/paper/2607.28595/) (2026)
- [VC-Tooler: Learning Compositional and Adaptive Visual Tool Use](https://mlsanity.com/paper/2608.02217/) (2026)
- [When Should a VLM Look? Paying Only for Visual Calls That Were Needed and Used](https://mlsanity.com/paper/2609.22910/) (2026)
- [VLM-in-Sandbox: Visual Workspaces for Agentic Visual Reasoning](https://mlsanity.com/paper/2609.24362/) (2026)
- [Rethinking Multi-Image Re-Representation in Multi-Image Understanding](https://mlsanity.com/paper/2609.39363/) (2026)
- [VisualCoding: Learning Executable Skills for Reliable Visual Reasoning](https://mlsanity.com/paper/2610.14403/) (2026)
- [Self-Evolving Visual Agent via Margin Reward](https://mlsanity.com/paper/2610.15084/) (2026)

## Abstract (arXiv)

Reinforcement learning for agentic multimodal models often suffers from interaction collapse, where models learn to reduce tool usage and multi-turn reasoning, limiting the benefits of agentic behavior. We introduce PyVision-RL, a reinforcement learning framework for open-weight multimodal models that stabilizes training and sustains interaction. Our approach combines an oversampling-filtering-ranking rollout strategy with an accumulative tool reward to prevent collapse and encourage multi-turn tool use. Using a unified training pipeline, we develop PyVision-Image and PyVision-Video for image and video understanding. For video reasoning, PyVision-Video employs on-demand context construction, selectively sampling task-relevant frames during reasoning to significantly reduce visual token usage. Experiments show strong performance and improved efficiency, demonstrating that sustained interaction and on-demand visual processing are critical for scalable multimodal agents.