100
Task Domains
50 / 50
ID / OOD
5
Samples/Task
Filter by Category:
🔍
G-131 · Select Next Figure Increasing Size Sequence
Samples:
00
01
02
03
04
1/5
PROMPT
Loading prompt...
GROUND TRUTH
First
First Frame
Final
Final Frame
Image Trajectory
Ground Truth

VBVR-Pro Bench

The official access-controlled release contains 100 tasks, split into 50 in-domain and 50 out-of-domain tasks, with five paired image/video cases per task. This local explorer indexes all 500 real cases.

Access-Controlled Bench VBVR EvalKit Foundation

Verifiable Protocol

Task-specific deterministic scorers evaluate the relevant final-state evidence. The Bench uses the same task/sample browser structure as VBVR. The aggregate Overall, ID, and OOD results below reproduce the values released on official public model cards.

VBVR-Pro Leaderboard

Released aggregate rankings for the 100-task VBVR-Pro Bench. Per-faculty columns are withheld until the public model cards and final paper use one frozen table.

Modality
Split
Type
# Model Type Modality Overall In-Domain Out-of-Domain
1
VBVR-Pro-Wan2.2-I2V-A14B
Pro Baseline Video 0.670 0.808 0.532
2
VBVR-Pro-SenseNova-U1
Pro Baseline Interleaved 0.638 0.811 0.464
3
SenseNova-U1-8B-MoT
Open Source Interleaved 0.565 0.533 0.597
4
Nano Banana Pro
Proprietary Interleaved 0.564 0.480 0.648
5
VBVR-Pro-Wan2.1-I2V-14B
Pro Baseline Video 0.562 0.730 0.395
6
Seedream-5.0-Pro
Proprietary Image 0.557 0.485 0.629
7
VBVR-Wan2.2
VBVR Baseline Video 0.517 0.548 0.486
8
GPT-Image-2
Proprietary Interleaved 0.507 0.428 0.587
9
SeedDance 2.0
Proprietary Video 0.499 0.451 0.547
10
VBVR-Pro-Wan2.2-TI2V-5B
Pro Baseline Video 0.470 0.641 0.300
11
VBVR-Pro-LTX2.3
Pro Baseline Video 0.425 0.527 0.324
12
VBVR-SenseNova-U1
VBVR Baseline Interleaved 0.408 0.469 0.347
13
VBVR-Pro-FLUX.2
Pro Baseline Image 0.407 0.484 0.330
14
Kling V3
Proprietary Video 0.392 0.356 0.427
15
VBVR-Pro-ThinkMorph
Pro Baseline Interleaved 0.373 0.402 0.344
16
VBVR-Pro-Qwen-Image
Pro Baseline Image 0.322 0.332 0.311
17
Qwen-Image-2.0
Proprietary Image 0.313 0.248 0.378
18
Veo 3.1
Proprietary Video 0.309 0.312 0.305
19
Wan2.2-I2V-14B
Open Source Video 0.182 0.157 0.207
20
VBVR-Pro-BAGEL
Pro Baseline Image 0.172 0.168 0.176
21
FLUX.2-dev
Open Source Image 0.157 0.108 0.206
22
ThinkMorph-7B
Open Source Interleaved 0.154 0.113 0.195
23
Qwen-Image-Edit
Open Source Image 0.134 0.108 0.159
24
LTX2.3-I2AV
Open Source Video 0.112 0.106 0.119
25
Wan2.1-I2V-14B
Open Source Video 0.100 0.105 0.095
26
Wan2.2-TI2V-5B
Open Source Video 0.094 0.066 0.122
27
BAGEL-7B-MoT
Open Source Image 0.089 0.066 0.111
28
CogVideoX1.5-5B-I2V
Open Source Video 0.085 0.100 0.070
29
HunyuanVideo-I2V
Open Source Video 0.054 0.054 0.053
29 / 29 released model results Source: official Video-Reason public model cards · accessed 2026-08-24
Official Results View Model Zoo