Interactive video generation and editing are becoming increasingly important for creative design. In this report, we introduce EditStream: a unified framework for interactive video generation and editing. EditStream unifies multiple video creation and manipulation tasks within a single DiT-based model through flexible task-specific conditioning, and further transforms it into a fast, few-step autoregressive model for efficient streaming. It supports Text-to-Video, Image-to-Video, Video-to-Video, Editing Propagation, Reference-guided Video Editing, and Camera Pose Change, enabling flexible control over video generation, transformation, and editing within one system. To make the unified model practical for interactive use, we develop a two-stage distillation approach that combines Velocity Moment Matching (VMM) with autoregressive unrolling. VMM matches conditional velocity moments at student-reached intermediate states to preserve generation quality and motion, while unrolling exposes the student to its own autoregressive predictions to improve temporal stability. Together, they alleviate common challenges in few-step autoregressive video generation, including over-saturation, degraded motion, temporal instability, and complex training. EditStream provides a practical and scalable solution that bridges high-quality diffusion-based video models with interactive creative workflows.
Figures & tables
Figure 1: EditStream adopts a dual-type conditioning interface for unified video generation and editing: pixel-aligned conditions (e.g., source latent video or camera-warped inputs y , and mask ϕm(m) ) are injected via compact channel concatenation x~t=Concatch[xt,y,ϕm(m)] to avoid increasing attention length, while non-aligned reference signals (e.g., a reference image or the first frame latent r for I2V/refV2V/EditProp) are provided by sequence token concatenation xt∗=Concattk[r,xt[:−1]] to better leverage self-attention for in-context learning.
Task
Abbrev.
Conditional Inputs
Special Task Tokens
Text-to-Video
T2V
Text prompt
–
Image-to-Video
I2V
First-frame image; (optional) Text prompt
–
(Egocentric) Video-to-Video
(Ego)V2V
Source (general/egocentric) video; Text prompt
–
Editing Propagation
EditProp
Source video; First-frame edit; (optional) Text prompt
[Editing Propagation]
Reference-guided Video-to-Video
refV2V
Source video; Reference image; Text prompt
[Reference Guidance]
Explicit Camera Pose Control
ReShoot
Re-projected source video; Mask; (optional) Text prompt
[Camera Control]
Table 1: Unified task formulation in EditStream. All tasks share the same video diffusion backbone and generate a video; they differ only in their conditional inputs and special task tokens.
Figure 2: Overview of VMM-forcing unrolling . During training, we fully unroll streaming inference: the causal student generates each temporal block from noise with the target few-step schedule, commits the denoised block to the KV cache as history, and proceeds to the next block. At a sampled local transition for each selected block/rollout, we query the bidirectional teacher and auxiliary model at the student-induced states, and supervise the student by matching the teacher–auxiliary residual velocity moments under the same autoregressive context. The student-generated first-block condition is also fed into the teacher and auxiliary model.
Category
Source Family
Scale
Sampling Ratio
Video Generation Data
Text-to-Video
LTX2.3-video, OpenVid-1M [ 50 ]
205K + 1.02M ( ∼ 1.23M clips)
21.875%
Text/Image-to-Video
LTX2.3-video, OpenVid-1M [ 50 ]
205K + 1.02M ( ∼ 1.23M clips)
21.875%
Video Editing and Control Data
Image Editing / Propagation
Pico-style image editing [ 54 ]
∼ 328K image pairs
9.375%
General Video Editing / Propagation
Curated paired video editing
∼ 37K pairs
6.25%
Table 2: Training Data Mixture. Sources, dataset scales, and sampling ratios used to train EditStream.
Table 3: Text-to-Video Results: VBench text-to-video benchmark. Latency and throughput are both end-to-end performance. Time-to-first-output denotes the wall-clock time until the first decoded RGB block becomes available at batch size 1. Measurements are conducted in one single H100 GPU.
Figure 3: Qualitative Text-to-Video (T2V) results on frequent user prompts. We compare the Self-forcing conversion and the proposed VMM-forcing pipeline. VMM-forcing produces richer motion dynamics and scene evolution while maintaining texture realism and temporal coherence, whereas Self-forcing often yields over-sharpened, plastic-like appearances and reduced temporal diversity.
Model
CLIP Temp. ↑
DINO Temp. ↑
Frame-Text Align. ↑
Video-Text Align. ↑
Pick Score ↑
Video Quality ↑
Commercial Models
Runway Aleph
0.9888
0.9849
28.30
25.33
20.36
7.64
Kling-O1
0.9926
0.9911
28.24
25.72
20.59
7.93
Offline Models
InsV2V [ 12 ]
0.9720
0.9679
25.84
23.07
19.55
5.07
LucyEditDev [ 15 ]
0.9852
0.9839
26.30
23.51
19.54
5.87
Table 4: Video-to-Video Editing : Comparison on EditVerse benchmark for video-to-video editing. Higher values indicate better performance for all metrics. We follow EditVerse to use GPT-4o as the VLM judge model.
Figure 4: Video-to-Video (V2V) editing results on EditVerse Benchmark compared with other streaming video editing baselines. Across a wide range of editing tasks, including object addition, removal, replacement, style transfer, text rendering, and complex compositional edits requiring precise pixel alignment and instruction following, EditStream consistently exhibits stronger semantic understanding, higher generation fidelity, and better color quality.
Model
Subject Reference
Background Reference
Overall ↑
Identity Consist. ↑
Temporal Consist. ↑
Physical Consist. ↑
Subj. Avg. ↑
Reference Similarity ↑
Matting Quality ↑
Video Quality ↑
BG Avg. ↑
Commercial Models
Runway Aleph
3.79
3.65
3.58
3.67
3.33
2.81
2.58
2.91
3.29
Kling-O1
4.75
4.66
4.60
4.67
3.95
3.21
2.75
3.30
3.99
Offline Models
UniVideo [ 68 ]
4.17
3.79
3.59
3.85
3.13
2.55
2.28
2.65
3.44
Table 5: Reference-Guided Video-to-Video Editing : Comparison on RefVIE-Bench with Gemini-3-Flash as the judge model.
Model
Subj. Cons. ↑
Back. Cons. ↑
Temp. Flicker ↑
Motion Smooth. ↑
Valid SSIM ↑
Hole Complete. ↑
Hole Temp. Stab. ↑
Gated Hole Fill Qual. ↑
Offline Models
ReCamMaster [ 1 ]
89.23
91.18
96.44
98.65
0.388
96.96
82.07
22.96
ViewCrafter [ 80 ]
86.86
91.31
93.26
96.12
0.407
96.37
86.43
27.73
GCD [ 64 ]
84.03
90.42
92.39
95.87
0.315
96.60
83.71
21.72
TrajCrafter [ 79 ]
88.19
92.13
93.49
97.00
0.661
94.89
87.54
39.06
Vista4D [ 40 ]
88.81
92.28
93.22
96.77
0.751
96.28
86.26
51.25
Table 6: Camera Pose Control : Comparison on the V2VCameraPose benchmark (472 samples). The first four columns are VBench dimensions computed purely from the video itself (no prompt/text dependence); the last four evaluate the quality of content hallucinated into the disoccluded (never-observed) regions produced by reprojecting each sample’s monocular point cloud onto the target camera trajectory, and how faithfully the observed (point-cloud-hit) region is preserved. Higher values indicate better performance for all metrics.
Figure 5: Reference Video-to-Video (refV2V) results on RefVIE Benchmark. Qualitatively, EditStream (online streaming model) consistently outperforms Kiwi-Edit (offline model) in preserving the reference subject while generating more realistic edited videos. Compared with the baseline, it produces more faithful identity preservation, more appropriate object scale, more coherent scene composition with the reference background, and richer pixel-level details.
Figure 6: Camera Pose (ReShoot) Control Results on the collected V2VCameraPose Benchmark. TrajCrafter and Vista4D are both strong offline baselines. However, they occasionally suffer from over-hallucination, introducing implausible objects or environmental elements. EditStream generally produces milder distortions while avoiding unnecessary hallucinated content.
Figure 7: Egocentric Editing (Ego V2V) Results. EgoEdit tends to produce sharper outputs, but they often appear less realistic. It may partly reflect the sharpness and mode-seeking behavior associated with DMD-based training. In the dog editing example, EgoEdit generates overly pronounced fur textures and introduces unrealistic specular highlights on the dog’s body. Similarly, in the lava editing example, the generated lava shows a CGI-like texture. EditStream with VMM produces outputs that are slightly less sharp but exhibit more faithful material properties, more realistic textures, and a more natural overall appearance.
Figure 8: Editing Propagation Results. AnyV2V and Señorita-2M struggle in challenging scenarios. Although EditStream is not trained specifically for outpainting, it demonstrates strong generalization capability. EditStream results are more realistic and detailed than that of GenProp (e.g. it produces a more realistic effect behind the car). It may partly benefit from differences in the underlying foundation model.
Figure 9: Editing Propagation Results for human-object interaction cases. Without the proposed HOI training data, the model struggles to establish stable physical correspondence between hands and edited objects. As a result, the generated object frequently disappears, fails to follow the intended manipulation, or exhibits severe geometric morphing during interaction. In contrast, incorporating HOI supervision significantly improves the realism and controllability of human–object interactions, making edited virtual objects behave naturally under user manipulation.
Figure 16
Configuration
Results
Warm-up
Unrolling
T2V
V2V
RefV2V
ID
Method
Iter.
W/U Ratio
Method
Alt.
Commit
Quality
Semantic
Overall
VLM
Overall
B1
VMM Init.
200
1:4
VMM-forcing Unrolling
✓
✓
85.36
80.64
84.42
7.35
3.71
B2
ODE Init.
5K
5:1
Self-forcing Unrolling
N/A
N/A
84.09
81.95
83.67
7.01
3.18
E1
ODE Init.
5K
1:0
N/A
N/A
N/A
81.72
74.33
80.24
6.98
2.68
E2
VMM Init.
200
1:0
N/A
N/A
N/A
84.17
79.89
83.31
7.13
3.55
Table 8: Ablation study on the proposed training strategy. Higher is better for all metrics.
Figure 11: Ablation study visualizations for warm-up existence and initialization methods.
Appendix figures & tables1 asset
Supplementary material from the paper’s appendix.
Appendix
Teacher
Student (ours)
Self-forcing baseline
Init
Wan2.2-TI2V-5B pretrained
Teacher EMA weights
Causal ODE-regression ckpt
Causal / block size
No
Yes, 3 frames/block
Yes, 3 frames/block
Denoising steps
full schedule
4 ({999,938,833,625})
4 ({999,938,833,625})
Distillation loss
flow matching
VMM
DMD
Block unrolling
–
yes (200-step warmup)
yes
Critic:generator ratio
–
1:1
5:1
Appendix
Table 9: Implementation summary across the three training stages.
Although existing video editing methods are generally feasible, they often require many costly iterations and still struggle to deliver high-quality yet satisfying editing results. We attribute this limitation to the prevalent data-to-data paradigm, which is less compatible with modern generative models than noise-to-data generation. To address this gap, we revisit video editing from a noise-to-data perspective and propose Streaming-Generation-based Video Editing (StreamEdit), which preserves few-step sampling while seamlessly injecting source-video conditions. Built on pre-trained streaming generation models, StreamEdit introduces dual-branch fast sampling with a self-attention bridge and cross-attention grounding/boosting to satisfy both sampling and conditioning requirements. We further propose source-oriented guidance to improve target-generation quality, and a visual prompting strategy to enhance editing flexibility and practicality. The method is effective, robust, and generalizable across different models. Extensive experiments on diverse video editing tasks show that StreamEdit consistently outperforms existing approaches, even in few-step settings with minimal time cost. Code and results are available at: https://dsl-lab.github.io/StreamEdit/.
Guanlong Jiao, Chenyangguang Zhang, Jia Jun Cheng Xian +2
The University of British Columbia · Vector Institute · ETH Zürich +2
Real-time video editing requires low-latency causal generation with bounded computational resources while preserving source fidelity and long-term temporal consistency. We present JoyAI-Video-Edit, a 16B-parameter autoregressive diffusion framework for real-time, open-ended video editing without access to future frames or a predefined video duration. Our method combines chunk-wise autoregressive adaptation, Source-Anchored Distribution Matching Distillation (SA-DMD), and Long-Horizon Autoregressive Distillation to reduce train--inference mismatch, preserve source fidelity during two-step generation, and mitigate accumulated temporal drift. Extensive automatic and human evaluations show that JoyAI-Video-Edit substantially outperforms existing streaming editors and remains competitive with strong offline systems on both short and long videos. The complete system achieves end-to-end 720p video editing at approximately 30 FPS on a single Nvidia B200 GPU. Code is available at https://github.com/jd-opensource/JoyAI-Video-Edit.
Streaming video editing has made rapid progress, yet practical deployment is still limited by two core issues: maintaining stable backgrounds and non-edited regions over time, and achieving the low latency required for real-time interactive scenarios. Meanwhile, recent streaming video generation methods are mostly developed for synthesis and cannot be directly applied to editing due to the strict preservation requirement and region-specific control. In this work, we present a novel streaming video editing framework that performs causal, frame-by-frame editing with strong content preservation and real-time responsiveness. Our key design is a three-stage distillation pipeline that progressively transfers editing capability from a powerful bidirectional foundation model to an efficient unidirectional streaming editor, enabling stable long-horizon edits without sacrificing visual fidelity. To further support real-time deployment, we introduce an AR-oriented mask cache that reuses region-related computation across frames, substantially reducing redundant processing and accelerating inference. Finally, we establish a dedicated benchmark for streaming video editing. Extensive evaluations demonstrate that our method achieves state-of-the-art visual quality among streaming baselines while drastically boosting inference speed to 12.66 FPS, making it suitable for interactive and augmented reality applications.