UnifiedReward

[ECCV 2026] UniREditBench: A Unified Reasoning-based Image Editing Benchmark: We propose UniREditBench, a unified reasoning-based image editing benchmark, and further construct UniREdit-Data-100K, a large-scale synthetic dataset with high-quality CoT annotations, and develop UniREdit-Bagel by fine-tuning Bagel on this dataset.

UniGenBench++: A Unified Semantic Evaluation Benchmark for Text-to-Image Generation: We propose UniGenBench++, a unified semantic benchmark for T2I generation. It supports both short and long prompts in Chinese and English, featuring a streamlined evaluation pipeline and a robust offline evaluation model.

Unified Personalized Reward Model for Vision Generation: We propose UnifiedReward-Flex, a unified personalized reward model for vision generation that couples reward modeling with flexible and context-adaptive reasoning.

Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning: We propose Pref-GRPO and UniGenbench, the first preference reward-based GRPO method for stable T2I reinforcement learning, and a unified T2I generation benchmark for fine-grained semantic consistency evaluation.

[NeurIPS 2025] Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning: We propose UnifiedReward-Think, the first unified multimodal CoT reward model.

Unified Reward Model for Multimodal Understanding and Generation: We release the UnifiedReward, the first unified reward model for multimodal understanding and generation assessment, enabling both pairwise ranking and pointwise scoring.

UnifiedReward Model Zoo

Model Name	Version	Inference Code Path	Info
UnifiedReward-Flex	Qwen35 / Qwen3-VL	Qwen35 / Qwen3-VL	Unified personalized reward model for vision generation.
UnifiedReward-Edit	Qwen35 / Qwen3-VL / Qwen25-VL	Qwen35 / Qwen3-VL / Qwen25-VL	Unified image generation & editing reward, trained on UnifiedReward-2.0.
UnifiedReward-2.0	Qwen35 / Qwen3-VL / Qwen25-VL	New / Classic Capabilities	New capabilities added for image & video generation reward.
UnifiedReward-Think	Qwen35 / Qwen3-VL / Qwen25-VL / LLaVA	Qwen35 / Qwen3-VL / Qwen25-VL / LLaVA	CoT integration to all vision reward tasks.
UnifiedReward	Qwen25-VL / LLaVA	Qwen25-VL / LLaVA	Unified image&video und./gen. reward models.

✨ Awesome Works using UnifiedReward

Meta, Transition Matching: Scalable and Flexible Generative Modeling.

Meta, Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image.

NVIDIA, Stanford, Tsinghua, DiffusionNFT: Online Diffusion Reinforcement with Forward Process.

PKU, ByteDance Seed, Enhancing Spatial Understanding in Image Generation via Reward Modeling.

Apple, Fudan, UniGen-1.5: Enhancing Image Generation and Editing through Reward Unification in Reinforcement Learning.

ByteDance, MammothModa2: A Unified AR–Diffusion Framework for Multimodal Understanding and Generation.

NPU, PKU, CUHK, Are We Ready for RL in Text-to-3D Generation? A Progressive Investigation.

University of California, USTC, PKU, BIGAI, MILR: Improving Multimodal Image Generation via Test-time Latent Reasoning.

Kuaishou, Tsinghua, CUHK, Flow-GRPO: Training Flow Matching Models via Online RL.

Tencent Hunyuan, MixGRPO: Unlocking Flow-based GRPO Efficiency with Mixed ODE-SDE.

Kling Team, CUHK MMLab, NJU, VR-Thinker: Boosting Video Reward Models through Thinking-with-Image Reasoning.

CUHK MMLab, Delving into RL for Image Generation with CoT: A Study on DPO vs. GRPO.

Method	HPS	ImageReward	UnifiedReward
Janus-Pro + DPO	77.3	77.7	80.0
Janus-Pro + GRPO	79.2	79.3	81.0
Janus-Pro + Best-of-4	82.1	82.4	84.5

Tencent Hunyuan X, X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Again.

News

[2026/3/19] We release UnifiedReward-Think-qwen35 and UnifiedReward-Flex-qwen35!!!

[2026/3/7] We release UnifiedReward-2.0-qwen35 and UnifiedReward-Edit-qwen35!!!

[2026/2/01] We release UnifiedReward-Flex-qwen3vl and its training data UnifiedReward-Flex-SFT-90K. The inference code is provided at here.

[2025/11/17] We release UnifiedReward-Think-qwen3vl-[2b/4b/8b/32b]. The inference code is provided at here.

[2025/11/11] We release UnifiedReward-2.0-qwen3vl-[2b/4b/8b/32b] and UnifiedReward-Edit-qwen3vl-[2b/4b/8b/32b]!!!

[2025/10/23] We release UnifiedReward-Edit-qwen-[3b/7b/32b/72b], a unified reward model for both Text-to-Image and Image-to-Image generation trained on approximately 700K unified image generation and editing reward data!! For image editing reward task, our models support:

1. Pairwise Rank — directly judge which of two edited images is better.

2. Pairwise Score — assign a separate score to each image in a pair.

3. Pointwise Score — rate a single image on two axes: instruction-following and overall image quality.

The image editing reward inference code is available at UnifiedReward-Edit/ directory, while T2I inference code is unchanged from previous models. The editing training data is preprocessed from EditScore and EditReward and will be released soon. We sincerely appreciate all contributors!!

[2025/9/25] We release UnifiedReward-2.0-qwen-[3b/[7b](https://huggingface.co/CodeGoat24/UnifiedReward-2.0-q