AI Paper Atlas

Research map / Benchmarks / HR-Bench 4K

HR-Bench 4K leaderboard

16 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on HR-Bench 4K. Best accuracy: 86.7 (GPT-5-mini + InSight-o3-vS, ICLR 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

HR-Bench 4K · accuracy

16 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
186.7GPT-5-mini + InSight-o3-vS
GPT-5-mini
InSight-o3: Empowering Multimodal Foundation Models with Generalized Visual Search
ICLR 2026 · Table 1
Multimodal LLMs
284.5FOVEA
Qwen3-VL-30B-A3B-Instruct
The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental Design
ICML 2026 · Table 1
Multimodal LLMs
380.1ARM-Thinker-7B
Qwen2.5-VL-7B
ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning
CVPR 2026 · Table 2
Multimodal LLMs
478.1PyVision-Image
Qwen2.5-VL-7B
PyVision-RL: Forging Open Agentic Vision Models via RL
ICML 2026 · Table 1
Multimodal LLMs
577.9DeepEyesV2
Qwen2.5-VL-7B
DeepEyesV2: Toward Agentic Multimodal Model
ICLR 2026 · Table 2
Multimodal LLMs
677.5Mini-o3 (Ours)
Qwen2.5-VL-7B-Instruct
Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search
ICLR 2026 · Table 1
Multimodal LLMs
776.6VisionLeaf
Qwen2.5-VL-7B
VisionLeaf: Entropy-Guided Leaf-First Reasoning for Efficient and Accurate Think-with-Image
CVPR 2026 · Table 1
Reasoning and chain of thought
875.6SV w/ Qwen2.5-VL-72B Verdict
Qwen2.5-VL-72B verdict
Small Drafts, Big Verdict: Information-Intensive Visual Reasoning via Speculation
ICLR 2026 · Table 1
Multimodal LLMs
975.6CodeVision-7B
Qwen2.5-VL-7B
Thinking with Programming Vision: Towards a Unified View for Thinking with Images
CVPR 2026 · Table 2
Multimodal LLMs
1075.2CodeDance-7B
Qwen2.5-VL-7B
CodeDance: A Dynamic Tool-integrated MLLM for Executable Visual Reasoning
CVPR 2026 · Table 1
Multimodal LLMs
6 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.