OpenCodePapers

spatial-reasoning-on-mmsi-bench

Visual ReasoningSpatial Reasoning
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeAccuracyModelNameReleaseDate
MJev-Spatial: System One Spatial Intelligence via Native Attention Matching✓ Link45.6MJev-Spatial2026-10-15
Scaling Spatial Intelligence with Multimodal Foundation Models45.2Gemini-3-Pro-Preview2025-11-17
Scaling Spatial Intelligence with Multimodal Foundation Models43.3SenseNova-SI (InternVL3-8B)2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence41.9GPT-52025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence41.0o32025-05-29
PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models✓ Link41.0PhysBrain 1.5-8B2026-09-14
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence40.3GPT-4.52025-05-29
MV-STRIDE: Enabling MLLMs to Master Multi-View Spatial Reasoning via Hierarchical Capability Modeling✓ Link38.9MV-STRIDE-SFT (Full-data SFT) (extra data)2026-09-08
Scaling Spatial Intelligence with Multimodal Foundation Models38.3Seed-1.62025-11-17
Scaling Spatial Intelligence with Multimodal Foundation Models38.1SenseNova-SI (Qwen3-VL-8B)2025-11-17
Scaling Spatial Intelligence with Multimodal Foundation Models37.8Grok-42025-11-17
Zooming without Zooming v2: From Seeing Details to Connecting Evidence for Generalizable Multimodal Perception✓ Link37.1ZwZ2-27B2026-10-27
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence36.9Gemini-2.5-Pro2025-05-29
RynnUnify: Towards Unified Embodied Brain✓ Link35.9RynnUnify2026-10-23
Visual Spatial Tuning34.8VST-7B-RL2025-11-07
Scaling Spatial Intelligence with Multimodal Foundation Models34.5SenseNova-SI (BAGEL-7B-MoT)2025-11-17
Scaling Spatial Intelligence with Multimodal Foundation Models34.2SenseNova-SI (InternVL3-2B)2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence34.1dots.vlm 12025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence33.8GLM-4.5V-thinking2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence33.0Doubao-1.5-pro2025-05-29
Visual Spatial Tuning32.0VST-7B-SFT2025-11-07
Visual Spatial Tuning31.3VST-3B-RL2025-11-07
Scaling Spatial Intelligence with Multimodal Foundation Models31.1Qwen3-VL-8B-Instruct2025-11-17
Scaling Spatial Intelligence with Multimodal Foundation Models31.0BAGEL-7B-MoT2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence30.9GPT-4.12025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence30.7Qwen2.5-VL-72B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence30.5NVILA-15B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence30.3GPT-4o2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence30.2Claude-3.7-Sonnet2025-05-29
Visual Spatial Tuning30.2VST-3B-SFT2025-11-07
Scaling Spatial Intelligence with Multimodal Foundation Models30.2ViLaSR-7B2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence29.7Seed1.5-VL2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence29.0InternVL2.5-2B2025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models28.9Qwen3-VL-2B-Instruct2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.7InternVL2.5-8B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.6DeepSeek-VL2-Small2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.5InternVL3-78B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.5InternVL2.5-78B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.4LLaVA-OneVision-72B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.1NVILA-8B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence28.0InternVL2.5-26B2025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models28.0InternVL3-8B2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence27.9InternVL2.5-38B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence27.7Qwen2.5-VL-32B2025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models27.4SpatialLadder-3B2025-11-17
Scaling Spatial Intelligence with Multimodal Foundation Models27.4SpaceR-7B2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence27.1DeepSeek-VL22025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models27.1Cambrian-S-7B2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence27.0InternVL3-1B2025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models27.0Cambrian-S-3B2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence26.8InternVL3-14B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence26.7InternVL3-9B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence26.5Qwen2.5-VL-3B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence26.3InternVL3-38B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence26.3InternVL2.5-4B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence26.1InternVL2.5-1B2025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models26.1Spatial-MLLM-4B2025-11-17
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence25.9Qwen2.5-VL-7B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence25.4Llama-3.2-11B-Vision2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence25.3InternVL3-2B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence24.5LLaVA-OneVision-7B2025-05-29
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence24.0DeepSeek-VL2-Tiny2025-05-29
Scaling Spatial Intelligence with Multimodal Foundation Models1.7MindCube-3B-RawQA-SFT2025-11-17
Check out how to contribute new results. Then edit this file on GitLab or GitHub.