Image editing

FLUX.1 Kontext · 2025-06Editing in the VAE latentIn-context editing

Handles generation and editing in one rectified-flow model by the simplest possible mechanism: concatenate the reference image's tokens onto the generation sequence and let attention do the rest. No adapters, no inversion, no task…

OminiControl · 2024-11Adapter-based conditioningIn-context editing

Predates FLUX.1 Kontext by about seven months and generalizes the same core move beyond editing: reuse the DiT's own VAE encoder and transformer blocks for the condition image, concatenate condition and generation tokens into one …

IP-Adapter · 2023-08Adapter-based conditioning

Moves adapter-based conditioning from spatial control to image-prompt conditioning: adds a second cross-attention branch with its own key/value projections just for image-prompt features, summed with the frozen model's existing te…

ControlNet · 2023-02Adapter-based conditioning

Freezes a pretrained UNet diffusion model entirely and adds a trainable copy of its encoder blocks, fed a spatial control signal (edges, depth, pose, segmentation) and connected back into the frozen backbone through zero-initializ…

InstructPix2Pix · 2022-11Editing in the VAE latent

Creates the instruction-editing task by manufacturing its training data: a language model writes edit instructions and a prompt-to-prompt pipeline renders the before and after images, giving supervised triplets that never existed.…

Prompt-to-Prompt · 2022-08Training-free attention manipulation

Observes that cross-attention maps bind words to spatial regions, then edits by recording the source prompt's attention maps and injecting, swapping or reweighting them under the edited prompt. No mask, no fine-tuning, no inversio…

HiDream-O1-Image · 2026-05Editing inside a unified modelSingle-stage pixel transformers

Maps raw image pixels, text tokens and task-specific condition tokens into one shared token space processed by the same transformer blocks through joint self-attention, eliminating the VAE and the separate text encoder entirely. E…

Qwen-Image 2.0 · 2026-05VAE-latent diffusion

The successor generation of Qwen-Image, extending fidelity and editing. Included because the Qwen-Image family is the open baseline most 2026 work measures against. Release status is unresolved: the technical report does not state…

RPiAE · 2026-03Editing in a representation latent

Initializes from a representation encoder but fine-tunes it, built explicitly to repair editing quality. Its abstract names the same bottleneck PS-VAE found: frozen-encoder tokenizers suffer limited reconstruction fidelity, which …

DiDAE · 2026-01Editing in a representation latent

The cleanest existing case of editing inside a genuinely frozen representation latent. It finds editing directions by Procrustes alignment and SVD directly in a frozen CLIP embedding space, transplanting the InterFaceGAN and GANSp…

PS-VAE · 2025-12Editing in a representation latent

Starts from DINOv2 and SigLIP2 features and does general instruction editing with a Transfusion-style transformer diffusing in a 96-channel latent. Its importance to this atlas is the reason it does not keep the encoder frozen: th…

Emu3.5 · 2025-10Editing inside a unified modelUnified understanding and generation

The closest thing to a genuinely single network among open unified models: one autoregressive transformer over interleaved text and vision tokens with one training objective. A discrete-diffusion adaptation module accelerates deco…

EditScore · 2025-09Reinforcement learning and preference alignment

Scores an edit as the geometric mean of semantic consistency, covering both instruction following and region preservation, and perceptual quality, validated against a companion benchmark and used to train a policy with Flow-GRPO.

Seedream 4.0 · 2025-09In-context editingVAE-latent diffusion

Folds text-to-image, editing, and multi-image composition into one system with native 4K output and dense text rendering. Representative of the industry-wide collapse of 'generation' and 'editing' into a single model rather than t…

OmniGen2 · 2025-06Editing inside a unified modelIn-context editingUnified understanding and generation

Uses two decoding pathways with unshared parameters for text and image, rather than forcing one shared path, and covers text-to-image, editing, and in-context generation in one model. A CVPR 2026 paper and the leading fully open u…

BAGEL · 2025-05Editing inside a unified modelUnified understanding and generation

A mixture-of-transformer-experts model with one expert for understanding and one for generation, routed over shared token sequences, trained on large-scale interleaved data. Unified at the sequence level, but not literally one net…

ImgEdit · 2025-05Evaluation and benchmarks

A large unified editing dataset and benchmark spanning many edit types, addressing the training-data bottleneck that instruction-editing quality is mostly limited by.

RISEBench · 2025-04Evaluation and benchmarks

The first benchmark for edits that require reasoning rather than recognition, across temporal, causal, spatial and logical cases. Its headline result is how far the field has to go: the strongest model tested reached only 35.9 per…

Step1X-Edit · 2025-04Editing in the VAE latentIn-context editing

An open instruction-editing system pairing a multimodal LLM that parses the instruction with a DiT that executes it, and the source of GEdit-Bench — the real-world editing benchmark most 2025-26 editing papers now report.

KV-Edit · 2025-02Training-free attention manipulation

The strictest preservation guarantee available: background tokens' key and value pairs are cached and held fixed, so only edit-region tokens regenerate while attending to the frozen background. Inversion-free with constant cache o…

RF-Solver · 2024-11Inversion for flow models

Attacks per-step numerical truncation error with an exact rectified-flow formulation and a high-order Taylor expansion in place of naive Euler steps. Its editing variant also shares self-attention features between the inversion an…

RF-Inversion · 2024-10Inversion for flow models

Reframes inversion for rectified-flow models as an optimal-control problem, equivalent to a rectified stochastic differential equation, so the trajectory is actively steered rather than passively re-integrated. Addresses drift, on…

GenArtist · 2024-07Agentic and search-augmented generation

A multimodal agent that plans a generation or edit, decomposes it into tool calls, and verifies the result before returning it. Bridges self-refinement loops to the current search-augmented wave.

InstantID · 2024-01Adapter-based conditioning

From a single face photo, preserves identity in seconds with no per-identity fine-tuning by imposing "strong semantic and weak spatial conditions" through a novel IdentityNet: facial embedding and landmark images are integrated wi…

RadEdit · 2023-12Medical: borrowed natural-image latent

Uses masked diffusion editing to synthesize the dataset shifts a deployed model will face, so a vision model can be stress-tested without collecting new clinical data. Editing as an evaluation instrument rather than as a content-c…

Emu Edit · 2023-11Editing in the VAE latent

Trains one editor across region-based edits, free-form edits and recognition tasks such as detection and segmentation recast as generation, steering between them with learned task embeddings. Generalizes poorly to instructions out…

BiomedJourney · 2023-10Medical: borrowed natural-image latent

Learns disease-progression editing from pairs of longitudinal radiology reports, using a language model to turn two visits into an edit instruction. Latent diffusion, and one of the few genuine image-editing systems in the medical…

MagicBrush · 2023-06Editing in the VAE latent

Over ten thousand human-annotated editing triplets where annotators were required to leave unrelated regions untouched, supplying the supervision synthetic pipelines could not. The paper's own conclusion is the useful one: even mo…

MasaCtrl · 2023-04Training-free attention manipulation

Converts self-attention into mutual self-attention: queries come from the target branch while keys and values are pulled from the source, so content is retrieved rather than resampled. A mask derived from cross-attention reduces f…

Plug-and-Play · 2022-11Training-free attention manipulation

Inverts the guidance image, then injects its recorded spatial convolutional features and self-attention maps into specific decoder layers and timesteps of the target generation. Preserves structure globally rather than by mask, so…

Pico-Banana-400K · 2025-10

400K edit pairs on real photographs, distilled from a frontier closed editor onto OpenImages. Notable both as Apple entering open editing data and as evidence that closed-model distillation is now a standard way to build open trai…

KRIS-Bench · 2025-05Evaluation and benchmarks

Critiques RISEBench as too coarse and not grounded in a formal knowledge framework, and replaces it with 1,267 instances across 22 tasks and seven reasoning dimensions, adding an explicit knowledge-plausibility metric.

ICEdit · 2025-04In-context editing

Shows in-context conditioning can replace heavy editing fine-tunes: state-of-the-art instruction editing with roughly 0.1% of the training data and 1% of the trainable parameters prior trained editors required.

ICE-Bench · 2025-03Evaluation and benchmarks

Evaluates creation and editing jointly, matching the architectural shift toward single models that do both.

ACE++ · 2025-01In-context editing

Unifies creation and editing under a long-context conditioning unit, implemented as a LoRA on FLUX.1 and therefore cheap to adapt. Worth noting that the paper itself reports no quantitative benchmark results, so its standing rests…

FireFlow · 2024-12Inversion for flow models

Targets the compute cost of precise inversion, claiming second-order solver accuracy at first-order cost with eight-step inversion and editing, roughly three times faster than comparable solvers. The three flow-inversion papers la…

Stable Flow · 2024-11Training-free attention manipulation

Diffusion transformers lack the UNet's coarse-to-fine hierarchy, so the question of which layer to inject into has no obvious answer. This paper finds the small set of vital layers that actually drive image formation and injects o…

MedEdit · 2024-07Medical: pixel space

Induces pathology into healthy brain MRI while preserving the surrounding scan. Notable for its evaluation: a board-certified neuroradiologist judged the generated strokes indistinguishable from real ones, which is a stronger bar …

PuLID · 2024-04Adapter-based conditioning

Tuning-free ID customization that pairs a standard diffusion branch with a parallel 'Lightning T2I' branch, adding a contrastive alignment loss and an accurate-ID loss so identity injection disturbs the base model's existing behav…

UniSpace · 2026-08Editing in a representation latent

An 8B mixture-of-experts unified understanding, generation and editing model built on a reparameterized frozen vision-transformer feature space rather than a VAE pathway. Flagged as needing an independent verification pass before …

RefEdit · 2025-06

Isolates a specific unsolved failure: when a scene has several similar objects, instruction editors cannot reliably pick the one the referring expression names.