AI Paper Atlas

Research map / Benchmarks / MJHQ-30K

MJHQ-30K leaderboard

18 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MJHQ-30K. Best FID: 1.46 (Infinity + LazyVAR, CVPR 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.

MJHQ-30K · FID

18 papers, each with its best reported number for its own method; lower is better.

#FID ↓MethodPaperTopic
11.46Infinity + LazyVAR
Infinity-2B
LazyVAR: Accelerating Visual Autoregressive Models via Scale-wise Token Pruning and Parallel Group Decoding
CVPR 2026 · Table 1
Autoregressive and latent generation
25.15Harmon-1.5B
Qwen2.5-1.5B-Instruct + MAR-H
Harmonizing Visual Representations for Unified Multimodal Understanding and Generation
ICCV 2025 · Table 3
Human and embodied modeling
35.28TiM 8-NFE
TiM 865M
Transition Models: Rethinking the Generative Learning Objective
CVPR 2026 · Table 2
Fast diffusion sampling
45.42SANA-1.5 4.8B Pre
SANA-1.0 1.6B
SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer
ICML 2025 · Comprehensive comparison table
Fast diffusion sampling
55.47Liquid-7BLiquid: Language Models are Scalable and Unified Multi-Modal Generators
IJCV 2026 · Table 3
Multimodal LLMs
65.5DIM-4.6B-T2I
Qwen2.5-VL-3B + SANA1.5-1.6B
Draw-In-Mind: Rebalancing Designer-Painter Roles in Unified Multimodal Models Benefits Image Editing
ICLR 2026 · Table 1
Text-to-image diffusion
75.6FUSEFUSE: Fine-Grained and Semantic-Aware Learning for Unified Image Understanding and Generation
AAAI 2026 · Table 2
Multimodal LLMs
85.9Gemma-2B-IT
T5-XXL
SANA: Efficient High-Resolution Text-to-Image Synthesis with Linear Diffusion Transformers
ICLR 2025 · Comparison of different Text-E
Text-to-image diffusion
96.05FOCUS
Qwen2.5-3B
FOCUS: Unified Vision-Language Modeling for Interactive Editing Driven by Referential Segmentation
NeurIPS 2025 · Table 2
Multimodal LLMs
106.1SynerGen-VL
InternLM2-1.8B
SynerGen-VL: Towards Synergistic Image Understanding and Generation with Vision Experts and Token Folding
CVPR 2025 · Table: Image generation result
Multimodal LLMs
8 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.