All General Coding Reasoning Multimodal Safety Specialized Agentic Language Long Context Speed Text-to-Image Image Editing Text-to-Video Image-to-Video
Image-to-Video Arena ELO (Artificial Analysis)76

Image-to-Video

Composite performance across 1 benchmarks · Normalized 0–100 · At least 1 scores required · Ranked by average performance · Top 50 of 76 eligible models

# Model Provider AA I2V Avg Coverage Price Value
1 Gemini Omni Flash Google 1,368
100.0
1/1 $0.10/s 986.2
2 Bach 1.0 Pro Video Rebirth 1,362
98.7
1/1 — —
3 Wan 3.0 Alibaba 1,362
98.7
1/1 $0.10/s 987.0
4 MiniMax H3 MiniMax 1,356
97.3
1/1 $0.13/s 748.5
5 Vimoo 1.0 Vimoo 1,348
95.5
1/1 — —
6 Dreamina Seedance 2.0 720p ByteDance Seed 1,339
93.5
1/1 $0.15/s 623.3
7 PixVerse V6 PixVerse 1,339
93.5
1/1 $0.040/s 2337.5
8 Grok Imagine Video 1.5 SpaceXAI 1,331
91.7
1/1 $0.14/s 655.0
9 Grok Imagine Video SpaceXAI 1,328
91.1
1/1 $0.070/s 1301.4
10 HappyHorse-1.1 Alibaba 1,314
87.9
1/1 $0.099/s 889.7
11 Kling 2.5 Turbo 1080p KlingAI 1,299
84.6
1/1 $0.070/s 1208.6
12 HappyHorse-1.0 Alibaba 1,294
83.4
1/1 $0.099/s 844.1
13 Vidu Q3 Pro Vidu 1,290
82.6
1/1 $0.10/s 826.0
14 PixVerse V5.6 PixVerse 1,287
81.9
1/1 $0.040/s 2047.5
15 Kling 3.0 Omni 1080p (Pro) KlingAI 1,281
80.5
1/1 $0.11/s 718.8
16 Kling 3.0 1080p (Pro) KlingAI 1,277
79.6
1/1 $0.11/s 710.7
17 PixVerse V5 PixVerse 1,277
79.6
1/1 $0.053/s 1493.4
18 Cosmos3-Super-Image2Video-4Step NVIDIA 1,275
79.2
1/1 — —
19 PixVerse V5.5 PixVerse 1,275
79.2
1/1 $0.053/s 1485.9
20 Kling 3.0 Omni 720p (Standard) KlingAI 1,271
78.3
1/1 $0.084/s 932.1
21 Wan 2.7 Alibaba 1,271
78.3
1/1 $0.10/s 783.0
22 Kling 3.0 720p (Standard) KlingAI 1,270
78.1
1/1 $0.084/s 929.8
23 Kling 2.6 Pro (January) KlingAI 1,266
77.2
1/1 $0.070/s 1102.9
24 Kling 2.6 Standard (January) KlingAI 1,265
77.0
1/1 $0.042/s 1833.3
25 Runway Gen-4.5 Runway 1,265
77.0
1/1 $0.12/s 641.7
26 Veo 3.1 Fast Google 1,259
75.6
1/1 $0.080/s 945.0
27 Hailuo 2.3 Fast MiniMax 1,257
75.2
1/1 $0.032/s 2372.2
28 Seedance 1.5 pro ByteDance Seed 1,256
74.9
1/1 $0.026/s 2880.8
29 Hailuo 2.3 MiniMax 1,255
74.7
1/1 $0.082/s 914.3
30 Veo 3.1 Google 1,253
74.3
1/1 $0.20/s 371.5
31 Cosmos3-Super-Image2Video NVIDIA 1,253
74.3
1/1 — —
32 Hailuo 02 Pro MiniMax 1,249
73.4
1/1 $0.082/s 898.4
33 Hailuo 02 Standard MiniMax 1,247
72.9
1/1 $0.047/s 1561.0
34 Wan 2.5 Preview Alibaba 1,245
72.5
1/1 $0.10/s 725.0
35 Veo 3.1 Lite Google 1,242
71.8
1/1 $0.030/s 2393.3
36 Vidu Q2 Turbo Vidu 1,237
70.7
1/1 $0.040/s 1767.5
37 Seedance 1.0 ByteDance Seed 1,235
70.2
1/1 $0.052/s 1363.1
38 Vidu Q2 Pro Vidu 1,233
69.8
1/1 $0.055/s 1269.1
39 Vidu Q3 Turbo Vidu 1,230
69.1
1/1 $0.055/s 1256.4
40 Veo 3 Google 1,225
68.0
1/1 $0.20/s 340.0
41 Wan 2.6 Alibaba 1,222
67.3
1/1 $0.10/s 673.0
42 LTX-2.5 Fast Lightricks 1,215
65.8
1/1 $0.090/s 731.1
43 Ray 3 Luma Labs 1,215
65.8
1/1 — —
44 Kling 2.1 Master KlingAI 1,208
64.2
1/1 $0.28/s 229.3
45 Hailuo 02 Fast MiniMax 1,207
64.0
1/1 — —
46 LTX-2 Pro Lightricks 1,201
62.6
1/1 $0.060/s 1043.3
47 Pika 2.5 Pika Art 1,199
62.2
1/1 — —
48 LTX-2.5 Pro Lightricks 1,194
61.1
1/1 $0.12/s 509.2
49 LTX-2 Fast Lightricks 1,191
60.4
1/1 $0.040/s 1510.0
50 Kling 2.1 Pro KlingAI 1,185
59.1
1/1 $0.098/s 603.1

Gemini Omni Flash leads on raw performance, but Seedance 1.5 pro delivers 192% more performance per dollar. That extra 25.1 points costs $0.003/s per point.

Access this data

JSON GET /api/v1
[
  {
    "rank": 1,
    "model": "Gemini Omni Flash",
    "provider": "Google",
    "avg_score": 100.0,
    "cost": 0.1014,
    "cost_unit": "video_second",
    "value": 986.2
  },
  {
    "rank": 2,
    "model": "Bach 1.0 Pro",
    "provider": "Video Rebirth",
    "avg_score": 98.7,
    "cost": null,
    "cost_unit": "video_second",
    "value": null
  },
  {
    "rank": 3,
    "model": "Wan 3.0",
    "provider": "Alibaba",
    "avg_score": 98.7,
    "cost": 0.1,
    "cost_unit": "video_second",
    "value": 987.0
  }
]
Curl
$ curl -s https://rightrank.ai/api/v1/rankings/image-to-video.json | jq '.[:3]'
CLI brew install rightrank
$ rightrank top image-to-video --limit 5
$ rightrank compare claude-opus-4.6 gpt-5.4-pro --category image-to-video
$ rightrank value image-to-video  # best performance per dollar
Claude Code Skill Skill
# In Claude Code, use the RightRank skill:
/rightrank best model for image-to-video
/rightrank compare top 5 by value
/rightrank cheapest model scoring above 80 on image-to-video