Skip to main content

GoogleVideo model

Veo 3.1

In text to video, Veo 3.1 is in place 10 of 23, with image to video in place 10. A 5 s clip costs about $2.00.

Image to videoRank 10 of 21

1,398Arena ratinghigher is better

Text to videoRank 10 of 23

1,364Arena ratinghigher is better

5 s clipRank 19 of 19

$2$0.40 per secondlower is better

Clip lengthRank 45 of 63

8slongest single cliphigher is better

PopularityRank 109 of 875

50out of 100same as last month

Summary

At $0.40 per second of video it is expensive; a typical video model costs $0.12.

New in Veo 3.1 are up to three reference images for consistent characters, objects, or scenes across shots. Google’s video model generates native audio, extends scenes, and supports first-to-last-frame transitions at up to 1080p and 8 seconds per clip.

Google schedules the veo-3.1-generate-preview endpoint for shutdown on October 22, 2026, and recommends Gemini Omni 1.1 Flash as its replacement.

Veo 3.1 benchmarks

Image to video arena

How people rate generated videos in blind comparisons, Elo points · higher is better

20 of 21 models

MiniMax H31,495
Gemini Omni 1.1 Flash1,488
Dreamina Seedance 2.5 720p1,477
Dreamina Seedance 2.0 720p1,475
Gemini Omni Flash1,465
Grok Imagine Video 1.51,456
FLUX.3 Video1,449
Wan 2.71,428
Grok Imagine Video1,415
Veo 3.11,398
Veo 3.1 Fast1,385
Vidu Q3 Pro1,363
Kling 3.0 Pro1,354
Wan 2.61,310
PixVerse V5.61,299
Kling 2.6 Pro1,294
Hailuo 2.31,262
HunyuanVideo 1.51,198
Wan 2.2 A14B1,170
LTX-2 19B1,159

as of September 22, 2026 · Source: LMArena

Text to video arena

How people rate generated videos in blind comparisons, Elo points · higher is better

20 of 23 models

Gemini Omni 1.1 Flash1,516
Gemini Omni Flash1,513
FLUX.3 Video1,493
Grok Imagine Video 1.51,492
Dreamina Seedance 2.0 720p1,479
Dreamina Seedance 2.5 720p1,474
MiniMax H31,460
Muse Video1,456
Sora 2 Pro1,368
Veo 3.11,364
Veo 3.1 Fast1,362
Sora 21,343
Grok Imagine Video1,342
Wan 2.71,337
Wan 2.61,327
PixVerse V5.61,240
Runway Gen-4.51,225
Kling 2.6 Pro1,216
Hailuo 2.31,206
HunyuanVideo 1.51,169

as of September 22, 2026 · Source: LMArena

Show all 188 individual results from 31 sources

Page 1 of 16

Published benchmarks

LMArena Text to Video

Elo ratings summarize pairwise comparisons within this leaderboard. They are not percentages. The scale does not express absolute quality or linear gaps between models.

Gemini Omni 1.1 Flash1,516 Elo
Gemini Omni Flash1,513 Elo
FLUX.3 Video1,493 Elo
Grok Imagine Video 1.51,492 Elo
Dreamina Seedance 2.0 720p1,479 Elo
Dreamina Seedance 2.5 720p1,474 Elo
MiniMax H31,460 Elo
Muse Video1,456 Elo
Sora 2 Pro1,368 Elo
Veo 3.11,364 Elo
Veo 3.1 Fast1,362 Elo
Sora 21,343 Elo
Grok Imagine Video1,342 Elo
Wan 2.71,337 Elo
Wan 2.61,327 Elo
PixVerse V5.61,240 Elo
Runway Gen-4.51,225 Elo
Kling 2.6 Pro1,216 Elo
Hailuo 2.31,206 Elo
HunyuanVideo 1.51,169 Elo
LTX-2 19B1,154 Elo
Wan 2.2 A14B1,132 Elo
Pika 2.21,009 Elo

higher is betterAs of November 5, 2026Scale: 0 Elo - 1,516 Elo

Source: LMArena
gradually.ai
Methodology and context
Metric
elo-rating
Scale
No upper bound is documented. The chart spans the recorded values and zero.
Participants
23
Sample
13,741
Source: LMArena

LMArena Image to Video

Elo ratings summarize pairwise comparisons within this leaderboard. They are not percentages. The scale does not express absolute quality or linear gaps between models.

MiniMax H31,495 Elo
Gemini Omni 1.1 Flash1,488 Elo
Dreamina Seedance 2.5 720p1,477 Elo
Dreamina Seedance 2.0 720p1,475 Elo
Gemini Omni Flash1,465 Elo
Grok Imagine Video 1.51,456 Elo
FLUX.3 Video1,449 Elo
Wan 2.71,428 Elo
Grok Imagine Video1,415 Elo
Veo 3.11,398 Elo
Veo 3.1 Fast1,385 Elo
Vidu Q3 Pro1,363 Elo
Kling 3.0 Pro1,354 Elo
Wan 2.61,310 Elo
PixVerse V5.61,299 Elo
Kling 2.6 Pro1,294 Elo
Hailuo 2.31,262 Elo
HunyuanVideo 1.51,198 Elo
Wan 2.2 A14B1,170 Elo
LTX-2 19B1,159 Elo
Pika 2.2997 Elo

higher is betterAs of November 5, 2026Scale: 0 Elo - 1,495 Elo

Source: LMArena
gradually.ai
Methodology and context
Metric
elo-rating
Scale
No upper bound is documented. The chart spans the recorded values and zero.
Participants
21
Sample
25,116
Source: LMArena

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

VVA-Bench attack success rate (no-defense baseline) (Show measurement, test conditions, and source)
Source value
74.8
Score
74.8
Metric
attack success rate (ASR)
Unit
%
Benchmark version
VVA-Bench Table 5, no-defense baseline
Category
image-to-video-safety
Direction
Lower is better
Sample
452
Source type
independent-evaluator
Evaluator
VVA-Bench, VPA-Guard paper Table 5
Status
active
Retrieved at
2026-10-17
Methodology
Independent evaluation. Image-to-video safety benchmark; baseline with no defense; exact Table 5 model column veo_31; 452 unsafe prompts; 45 safe-control prompts are separate. Output resolution not reported.
Context
ASR is attack success rate; the separate 45 safe controls are not included in this denominator.
Source VVA-Bench, VPA-Guard paper Table 5
VideoThinkBench (Show measurement, test conditions, and source)
Source value
27.7
Score
27.7
Metric
Average (vision-centric tasks, accuracy)
Unit
%
Benchmark version
VideoThinkBench mini test set, Table 7
Direction
Higher is better
Source type
independent-evaluator
Evaluator
Thinking with Video, VideoThinkBench (CVPR 2026 paper)
Status
active
Retrieved at
2026-10-19
Methodology
Independent evaluation. Paper-reported Vision-centric mini set; 500 dataset cases per Appendix A.2 row. VideoThinkBench mini test set; Appendix A.2 gives 750 cases in the mini test set, 500 vision-centric and 250 text-centric. This is not an established per-model denominator. No exact serving revision or model response count is reported.
Context
Multimodal reasoning with generated videos, not general video quality. The tables report no effective per-model denominator, confidence interval, or served revision. The aggregation definition for the Average table field is not further specified; the published value is retained unchanged.
Source Thinking with Video, VideoThinkBench (CVPR 2026 paper)
VideoThinkBench (Show measurement, test conditions, and source)
Source value
48.3
Score
48.3
Metric
Average (text-centric tasks, last-frame evaluation)
Unit
%
Benchmark version
VideoThinkBench mini test set, Table 8
Direction
Higher is better
Source type
independent-evaluator
Evaluator
Thinking with Video, VideoThinkBench (CVPR 2026 paper)
Status
active
Retrieved at
2026-10-19
Methodology
Independent evaluation. Paper-reported Text-centric mini set; last-frame evaluation; 250 dataset cases per Appendix A.2 row. VideoThinkBench mini test set; Appendix A.2 gives 750 cases in the mini test set, 500 vision-centric and 250 text-centric. This is not an established per-model denominator. No exact serving revision or model response count is reported.
Context
Multimodal reasoning with generated videos, not general video quality. The tables report no effective per-model denominator, confidence interval, or served revision. The aggregation definition for the Average table field is not further specified; the published value is retained unchanged.
Source Thinking with Video, VideoThinkBench (CVPR 2026 paper)
VideoThinkBench (Show measurement, test conditions, and source)
Source value
44.5
Score
44.5
Metric
Average (text-centric tasks, audio evaluation)
Unit
%
Benchmark version
VideoThinkBench mini test set, Table 8
Direction
Higher is better
Source type
independent-evaluator
Evaluator
Thinking with Video, VideoThinkBench (CVPR 2026 paper)
Status
active
Retrieved at
2026-10-19
Methodology
Independent evaluation. Paper-reported Text-centric mini set; audio answer evaluation; 250 dataset cases per Appendix A.2 row. VideoThinkBench mini test set; Appendix A.2 gives 750 cases in the mini test set, 500 vision-centric and 250 text-centric. This is not an established per-model denominator. No exact serving revision or model response count is reported.
Context
Multimodal reasoning with generated videos, not general video quality. The tables report no effective per-model denominator, confidence interval, or served revision. The aggregation definition for the Average table field is not further specified; the published value is retained unchanged.
Source Thinking with Video, VideoThinkBench (CVPR 2026 paper)
AA Video Arena Elo (Show measurement, test conditions, and source)
Source value
967
Score
967
Metric
pairwise human-preference Elo
Unit
elo
Benchmark version
AA-Video-T2V v2.1
Category
text-to-video
Direction
Higher is better
Sample
3043
Confidence interval
18
Lower confidence bound
949
Upper confidence bound
985
Source type
independent-evaluator
Evaluator
Artificial Analysis AA-Video-T2V v2.1
Status
superseded
Retrieved at
2026-10-16
Methodology
Independent evaluation. Text-to-video with audio; exact source row Veo 3.1.
Context
The published comparison includes audio; it is not a silent-video-only score.
Source Artificial Analysis AA-Video-T2V v2.1
AA Video Arena Elo (Show measurement, test conditions, and source)
Source value
1031
Score
1031
Metric
pairwise human-preference Elo
Unit
elo
Benchmark version
AA-Video-I2V v1.1
Category
image-to-video
Direction
Higher is better
Sample
3320
Confidence interval
17
Lower confidence bound
1014
Upper confidence bound
1048
Source type
independent-evaluator
Evaluator
Artificial Analysis AA-Video-I2V v1.1
Status
active
Retrieved at
2026-10-16
Methodology
Independent evaluation. Image-to-video with audio; exact source row Veo 3.1.
Context
The published comparison includes audio; it is not a silent-video-only score.
Source Artificial Analysis AA-Video-I2V v1.1
Video Models AI-judge benchmark (Show measurement, test conditions, and source)
Source value
3.77
Score
3.77
Metric
Weighted overall AI-judge quality score (0-5)
Unit
points
Benchmark version
Video Models leaderboard updated 2026-10-15
Category
video-generation
Direction
Higher is better
Sample
64
Source type
vendor-reported
Evaluator
Video Models by Wiro AI
Status
active
Retrieved at
2026-10-16
Methodology
Vendor-reported result. Exact Wiro API route google/veo3-1. Published method states one API run per prompt using model defaults, first successful output, a 5-second duration target, and the resolution closest to 720p. This route documentation lists 4, 6, or 8s; 720p or 1080p; maximum 8s. The CSV does not expose the exact per-prompt overrides or criterion-specific denominators.
Context
Wiro AI is a commercial model gateway. Scores use its weighted 0-5 AI-judge rubric, not a blind human pairwise preference panel. No per-criterion N or confidence interval is published. This is task-specific AI-judge evidence, not broad independent quality proof.
Source Video Models by Wiro AI
Video Models AI-judge benchmark (Show measurement, test conditions, and source)
Source value
3.44
Score
3.44
Metric
Prompt adherence AI-judge criterion score (0-5); rubric weight 21%
Unit
points
Benchmark version
Video Models leaderboard updated 2026-10-15
Category
video-generation
Direction
Higher is better
Source type
vendor-reported
Evaluator
Video Models by Wiro AI
Status
active
Retrieved at
2026-10-16
Methodology
Vendor-reported result. Exact Wiro API route google/veo3-1. Published method states one API run per prompt using model defaults, first successful output, a 5-second duration target, and the resolution closest to 720p. This route documentation lists 4, 6, or 8s; 720p or 1080p; maximum 8s. The CSV does not expose the exact per-prompt overrides or criterion-specific denominators.
Context
Wiro AI is a commercial model gateway. Scores use its weighted 0-5 AI-judge rubric, not a blind human pairwise preference panel. No per-criterion N or confidence interval is published. This is task-specific AI-judge evidence, not broad independent quality proof.
Source Video Models by Wiro AI
Video Models AI-judge benchmark (Show measurement, test conditions, and source)
Source value
3.65
Score
3.65
Metric
Motion and consistency AI-judge criterion score (0-5); rubric weight 19%
Unit
points
Benchmark version
Video Models leaderboard updated 2026-10-15
Category
video-generation
Direction
Higher is better
Source type
vendor-reported
Evaluator
Video Models by Wiro AI
Status
active
Retrieved at
2026-10-16
Methodology
Vendor-reported result. Exact Wiro API route google/veo3-1. Published method states one API run per prompt using model defaults, first successful output, a 5-second duration target, and the resolution closest to 720p. This route documentation lists 4, 6, or 8s; 720p or 1080p; maximum 8s. The CSV does not expose the exact per-prompt overrides or criterion-specific denominators.
Context
Wiro AI is a commercial model gateway. Scores use its weighted 0-5 AI-judge rubric, not a blind human pairwise preference panel. No per-criterion N or confidence interval is published. This is task-specific AI-judge evidence, not broad independent quality proof.
Source Video Models by Wiro AI
Video Models AI-judge benchmark (Show measurement, test conditions, and source)
Source value
3.75
Score
3.75
Metric
Camera and framing AI-judge criterion score (0-5); rubric weight 12%
Unit
points
Benchmark version
Video Models leaderboard updated 2026-10-15
Category
video-generation
Direction
Higher is better
Source type
vendor-reported
Evaluator
Video Models by Wiro AI
Status
active
Retrieved at
2026-10-16
Methodology
Vendor-reported result. Exact Wiro API route google/veo3-1. Published method states one API run per prompt using model defaults, first successful output, a 5-second duration target, and the resolution closest to 720p. This route documentation lists 4, 6, or 8s; 720p or 1080p; maximum 8s. The CSV does not expose the exact per-prompt overrides or criterion-specific denominators.
Context
Wiro AI is a commercial model gateway. Scores use its weighted 0-5 AI-judge rubric, not a blind human pairwise preference panel. No per-criterion N or confidence interval is published. This is task-specific AI-judge evidence, not broad independent quality proof.
Source Video Models by Wiro AI

Recorded measurement details

LMArena Text to Video: LMArena Text to Video (Show measurement, test conditions, and source)
Source value
1364
Score
1364
Metric
elo-rating
Unit
Elo
Task
lmarena-text-to-video
Task label
LMArena Text to Video
Direction
Higher is better
Source rank
12
Sample
13741
Lower confidence bound
1350
Upper confidence bound
1378
Retrieved at
2026-11-05
Source LMArena
LMArena Image to Video: LMArena Image to Video (Show measurement, test conditions, and source)
Source value
1398
Score
1398
Metric
elo-rating
Unit
Elo
Task
lmarena-image-to-video
Task label
LMArena Image to Video
Direction
Higher is better
Source rank
13
Sample
25116
Lower confidence bound
1388
Upper confidence bound
1408
Retrieved at
2026-11-05
Source LMArena

Page 1 of 16

What Veo 3.1 can do

“n/a” means the provider does not state it or we have not sourced it yet. That is not a “no”.

Text to videoyesSource
Image to videoyesSource
Referencesn/a
AudioyesSource
Runs locallynoSource
Extendn/a
Start and end framen/a
Max. resolution1080pSource

Video sample from our test

This video exists as a documented run for the exact model version. Playback is muted.

Motion and physical consistency

This sample tests Anatomy, object consistency, camera tracking, rain, and reflections.

Open video at original sizeOpen source
Prompt and settings

Generate exactly one continuous five-second 16:9 cinematic video. A fictional bicycle courier in a mustard-yellow raincoat rides a red city bicycle from left to right across a rain-wet Hamburg street at blue hour for the entire clip. The camera tracks her smoothly at waist height. Both hands remain visible on the handlebars, the bicycle geometry stays physically correct, and a transparent umbrella fixed to the bicycle moves naturally in the wind. Wet-street reflections and raindrops follow realistic physics. No cuts, no slow motion, no text, no logos, no subtitles, and no audio.

Generated through fal.ai with model fal-ai/veo3.1. Archived settings were 16:9, one request, first output, requested 6 s, seed provider random, request {"prompt":"Generate exactly one continuous five-second 16:9 cinematic video. A fictional bicycle courier in a mustard-yellow raincoat rides a red city bicycle from left to right across a rain-wet Hamburg street at blue hour for the entire clip. The camera tracks her smoothly at waist height. Both hands remain visible on the handlebars, the bicycle geometry stays physically correct, and a transparent umbrella fixed to the bicycle moves naturally in the wind. Wet-street reflections and raindrops follow realistic physics. No cuts, no slow motion, no text, no logos, no subtitles, and no audio.","aspect_ratio":"16:9","duration":"6s","resolution":"720p","generate_audio":false}, configured price estimate $1.20.

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

0 to 100, higher is more popular

20 of 98 models

LTX-2.570
MiniMax H366
Muse Video62
Sora 259
Wan 2.1 T2V 14B54
Wan 2.1 T2V 1.3B54
Runway Gen-4.553
Hy-Nomas53
Wan 2.1 I2V 14B53
Veo 3.150
Wan 2.2 TI2V 5B46
FLUX.3 Video45
LTX-2 19B45
Open-Sora 2.044
Wan 2.2 I2V A14B43
Gemini Omni Flash43
SkyReels V2 T2V 14B42
Grok Imagine Video42
Mochi 1 Preview39
Pika H3 Flex35

As of November 2026 · Source: gradually.ai popularity index

Over time

Weekly values from October 18, 2026 to November 5, 2026

Veo 3.1Runway Gen-4.5FLUX.3 VideoLTX-2 19BGemini Omni Flash

Veo 3.1 compared

Compare with another model

The most popular comparisons

  1. 1Sora 2 vs. Veo 3.1
  2. 2MiniMax H3 vs. Veo 3.1
  3. 3Veo 3.1 vs. Runway Gen-4.5
  4. 4Gemini Omni Flash vs. Veo 3.1
  5. 5Veo 3.1 vs. Grok Imagine Video 1.5
  6. 6Veo 3.1 vs. Kling 3.0 Pro
  7. 7Dreamina Seedance 2.5 720p vs. Veo 3.1
  8. 8Veo 3.1 vs. Grok Imagine Video 1.5 Lite
  9. 9FLUX.3 Video vs. Veo 3.1
Show all 48 comparisons
  1. 10Dreamina Seedance 2.0 720p vs. Veo 3.1
  2. 11Sora 2 Pro vs. Veo 3.1
  3. 12Veo 3.1 vs. Veo 3.1 Fast
  4. 13Veo 3.1 vs. Grok Imagine Video
  5. 14Veo 3.1 vs. Wan 2.7
  6. 15Veo 3.1 vs. Wan 2.6
  7. 16Veo 3.1 vs. Kling 2.6 Pro
  8. 17Veo 3.1 vs. Vidu Q3 Pro
  9. 18Veo 3.1 vs. PixVerse V5.6
  10. 19Veo 3.1 vs. Hailuo 2.3
  11. 20Veo 3.1 vs. LTX-2 19B
  12. 21Veo 3.1 vs. Wan 2.2 A14B
  13. 22Veo 3.1 vs. HunyuanVideo 1.5
  14. 23Veo 3.1 vs. Pika 2.2
  15. 24Veo 3.1 vs. Gemini Omni 1.1 Flash
  16. 25Veo 3.1 vs. Muse Video
  17. 26Veo 3.1 vs. Wan 3.0
  18. 27Veo 3.1 vs. Wan 3.0 Prime
  19. 28Veo 3.1 vs. HappyHorse 1.0
  20. 29Veo 3.1 vs. HappyHorse 1.1
  21. 30Veo 3.1 vs. PixVerse V6
  22. 31Veo 3.1 vs. PixVerse C1
  23. 32Veo 3.1 vs. Seedance 2.0 Fast
  24. 33Veo 3.1 vs. Seedance 2.0 Mini
  25. 34Veo 3.1 vs. Vidu Q3 Turbo
  26. 35Veo 3.1 vs. Vidu Q2
  27. 36Veo 3.1 vs. Vidu Q1
  28. 37Veo 3.1 vs. Kling 4.0
  29. 38Veo 3.1 vs. Kling 3.0 Standard API
  30. 39Veo 3.1 vs. Kling Video O1
  31. 40Veo 3.1 vs. Hailuo 2.3 Fast
  32. 41Veo 3.1 vs. Wan 2.5 Preview
  33. 42Veo 3.1 vs. Wan 2.6 I2V Flash
  34. 43Veo 3.1 vs. Wan 2.2 TI2V 5B
  35. 44Veo 3.1 vs. Wan 2.2 I2V A14B
  36. 45Veo 3.1 vs. HunyuanVideo 1.0 T2V
  37. 46Veo 3.1 vs. HunyuanVideo I2V 1.0
  38. 47Veo 3.1 vs. Ray 3.2
  39. 48Veo 3.1 vs. LTX-2.5

All specifications

Input and output

Maximum resolution
1080p
Source: Maximum resolution
Maximum duration
8 s
Source: Maximum duration
Native audio
Yes
Source: Native audio

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Subject to limitations. Google documents SynthID for videos from the former Gemini API preview endpoints. Those endpoints were shut down on October 22, 2026. Other access routes require separate verification.
Source: AI watermarking
Watermarking technique
SynthID
Source: Watermarking technique
Labelling reviewed
Nov 5, 2026
Source: Labelling reviewed

Licence and access

Weights
Proprietary
Source: Weights
Access
API, Web
Source: Access
Lifecycle
Generally available on Vertex AI as veo-3.1-generate-001 (retirement no earlier than November 17, 2026); the Gemini API preview veo-3.1-generate-preview was deprecated by Google for October 22, 2026
Source: Lifecycle

Prices

Choose a price view
Representative price
$0.40 per second ($2.00 per 5 s clip)Source
fal.ai (fal-ai/veo3.1)
$1.2 per request (estimated)Source

No Google subscription includes this model. Through other providers from ≈ €6.36 a month (ElevenLabs Starter, with limited credits). 68 plans in Germany.

Filter plans
PlanMonthly priceNote
Lovable Freefreewith limited credit
ElevenLabs Starter≈ €6.36with limited credit
Poe Basic€6.83with limited credit
Pika Starter≈ €10.60with limited credit
Adobe Firefly Standard€10.98with limited credit
Hailuo AI StandardMiniMax≈ €11.13with limited credit
Leonardo.Ai Essential≈ €12.72
Runway Standard≈ €15.90with limited credit
Magnific Premium€19.04with limited credit
Perplexity Pro≈ €21.20with limited credit
Recraft Pro≈ €21.20with limited credit
Adobe Firefly Pro€21.99with limited credit
ElevenLabs Creator≈ €23.33with limited credit
Recraft Teams≈ €23.33with limited credit, for teams, per seat
Lovable Pro€25with limited credit
Leonardo.Ai Team Starter≈ €25.45for teams, per seat
Poe Plus€27.36with limited credit
Leonardo.Ai Premium≈ €31.81
Luma Plus≈ €31.81with limited credit
Krea Pro≈ €37.11with limited credit
Pika Creator≈ €37.11with limited credit
Runway Pro≈ €37.11with limited credit
Hailuo AI ProMiniMax≈ €40.29with limited credit
Perplexity Enterprise Pro≈ €42.41with limited credit, for teams, per seat
Magnific Premium+€42.84with limited credit
Lovable Business€50with limited credit, for teams
Leonardo.Ai Team Growth≈ €50.89for teams, per seat
Adobe Firefly Pro Plus€54.99with limited credit
Mindverse Studio Team€58.31for teams, per seat
Leonardo.Ai Ultimate≈ €63.61
Poe Pro€68.41with limited credit
Higgsfield Plus€70.21with limited credit
Magnific Business€70.21with limited credit, for teams, per seat
Mindverse Studio Pro€70.21per seat
Runway Team≈ €73.16with limited credit, for teams, per seat
Higgsfield Team€94.01with limited credit, for teams, per seat
Hailuo AI MasterMiniMax≈ €94.36with limited credit
Luma Pro≈ €95.42with limited credit
Pika Fancy≈ €100.72with limited credit
Runway Max≈ €100.72with limited credit
Magnific Pro Starter€104.72with limited credit
ElevenLabs Pro≈ €104.96with limited credit
Krea Max≈ €111.32with limited credit
Poe Advanced€136.84with limited credit
Higgsfield Ultra€153.51with limited credit
Higgsfield Scale€201.11with limited credit, for teams, per seat
Krea Business≈ €212.05with limited credit, for teams
Perplexity Max≈ €212.05with limited credit
Adobe Firefly Premium€219.98with limited credit
Hailuo AI MaxMiniMax≈ €229.01with limited credit
ElevenLabs Scale≈ €317.01with limited credit, for teams
Luma Ultra≈ €318.07with limited credit
Poe Max€342.11with limited credit
Perplexity Enterprise Max≈ €344.57with limited credit, for teams, per seat
ElevenLabs Business≈ €1,049.63with limited credit, for teams
Adobe Firefly Premium für Teamsn/awith limited credit, for teams, per seat
Adobe Firefly Pro für Teamsn/awith limited credit, for teams, per seat
Adobe Firefly Pro Plus für Teamsn/awith limited credit, for teams, per seat
ElevenLabs Enterprisen/afor enterprises
Higgsfield Enterprisen/afor enterprises
Krea Enterprisen/afor enterprises
Leonardo.Ai Customn/afor enterprises
Lovable Enterprisen/afor enterprises
Luma Enterprisen/afor enterprises
Luma Teamn/afor teams
Magnific Enterprisen/afor enterprises
Pika Enterprisen/afor enterprises
Runway Enterprisen/afor enterprises

All plans in the AI subscription comparison

Frequently asked questions about Veo 3.1

How good is Veo 3.1?

Rank 10 of 23 in the text-to-video arena rating. With a start image: rank 10 of 21.

LMArena, as of Sep 22, 2026

Arena ratingtext to video, from user votes · higher is better
MiniMax H31,460
Muse Video1,456
Sora 2 Pro1,368
Veo 3.1Rank 10 of 231,364
Veo 3.1 Fast1,362
Sora 21,343
Grok Imagine Video1,342

How much does a video with Veo 3.1 cost?

About $2 for a 5 s clip, $0.40 per second of video. That is expensive.

Price per seconddollars per second of video · lower is better
Kling 2.6 Pro$0.14
Wan 2.7$0.15
Wan 2.6$0.15
Dreamina Seedance 2.0 720p$0.15
Kling 3.0 Pro$0.17
FLUX.3 Video$0.17
Veo 3.1Rank 19 of 19$0.40

How long can a video with Veo 3.1 be?

Up to 8 seconds in one go. Resolution: 1080p.

According to Google

Price
$0.40 per second$2.00 per 5 s clip
Longest clip
8s
Resolution
1080p
Audio
yes, generated natively
Open weights
no

Where can I use Veo 3.1?

Not in a Google subscription. Including other providers that makes 68 plans in Germany.

Subscription comparison, prices in Germany incl. VAT

At Google

No Google subscription includes this model.

Through other providers

68 plans from 15 providers, from ≈ €6.36 a month (ElevenLabs Starter), with limited credits.

Sources and data date

Show all 54 sources

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

EvaluatorTest seriesComposite indicesRecorded active measurements
Artificial Analysis004
LMArena004

Without vendor claims and historical or disputed results; those stay labelled in the single values. 8 measurements are not yet assigned to a test series.

TypeEvidence and data date
Research statusResearch date October 16, 2026. 45 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not found in the inspected sources

arena-video-edit-exact-row

Note

The inspected Video Edit table has no exact model-version row. T2V and I2V values were not reused.

Not found in the inspected sources

vabench-exact-model-row

Note

VABench has no exact model row. Sora 2 was not transferred to Sora 2 Pro or other models.

Unresolved

vbench-current-public-leaderboard

Note

The current VBench leaderboard is embedded in a dynamic Hugging Face iframe and was not retrievable; current scores remain unresolved.

Unresolved

vidbenchmark-performance-axes

Note

The source publishes speed-to-first-frame and cost-efficiency axes, but its inspected methodology does not define them as measured performance results or disclose units, direction, or a scoring formula. They are omitted. Model limits, maximum resolution, and list prices are not quality measurements.

Not published

AA and VVA output resolution

Note

The AA rows and VVA-Bench safety table do not publish model output resolution; CaliBench's Veo 3.1 row is explicitly 720p.

Unresolved

per-model-sample-count

Original wording

The returned row text contains an ambiguous count under headings that include both Responses and Matches. Per-model N is not copied. Aggregate votes and pairwise-match totals are not substituted.

Not published

provider-serving-revision

Original wording

The page names a model label but does not state an exact provider serving revision or checkpoint.

Unresolved

benchmark-update-date

Original wording

The page header says updated 2026-09-03 while the footer says last updated 2026-10-18.

Additional sourceGoogle · Google Developers Veo 3.1 announcement · Editorial description reviewed October 2, 2026
Additional sourceVideo pricing
Additional sourceLifecycle data
Additional sourcefal.ai
Additional sourceLMArena (retrieved November 5, 2026)12
Additional sourceArtificial Analysis AA-Video-T2V v2.1 (retrieved October 16, 2026; October 17, 2026; October 28, 2026)
Additional sourceArtificial Analysis AA-Video-I2V v1.1 (retrieved October 16, 2026; October 17, 2026; October 28, 2026)
Additional sourceVideo Models by Wiro AI (retrieved October 16, 2026; October 28, 2026)
Additional sourcevidbenchmark (retrieved October 16, 2026)
Additional sourceCaliBench, Table 2 (retrieved October 17, 2026)
Additional sourceVVA-Bench, VPA-Guard paper Table 5 (retrieved October 17, 2026)
Additional sourceDream.exe, arXiv:2606.04811v1, Table 4 (retrieved October 17, 2026)
Additional sourceSandBase API video benchmark, model/prompt score table (retrieved October 17, 2026)
Additional sourceHeyGen Avatar V technical report, Table 1 (retrieved October 17, 2026)
Additional sourceT2AV-Compass, Table 3 (retrieved October 17, 2026; October 22, 2026)
Additional sourceV2V-Bench, Tables 2 and 3 (retrieved October 17, 2026)
Additional sourceBRITE: A Benchmark for Reliable and Interpretable T2V Evaluation on Implausible Scenarios (retrieved October 17, 2026)
Additional sourceVWG-Bench, Table 2 (retrieved October 17, 2026)
Additional sourcePhyGround paper authors, arXiv v1 (retrieved October 17, 2026)
Additional sourceArena Text-to-Video leaderboard (retrieved September 21, 2026; October 16, 2026; October 17, 2026; October 28, 2026)
Additional sourceArena Image-to-Video leaderboard (retrieved September 21, 2026; October 16, 2026; October 17, 2026; October 28, 2026)
Additional sourceContra Labs Human Creativity Benchmark Appendix (retrieved October 17, 2026)
Additional sourceVIBE AI Video Model Benchmarks (retrieved October 17, 2026)
Additional sourceVGI-Bench paper (retrieved October 17, 2026)
Additional sourceAVGen-Bench paper (retrieved October 17, 2026)
Additional sourceSci-VBench paper, arXiv:2608.09873v1 (retrieved October 17, 2026)
Additional sourceRapidata (retrieved October 18, 2026; October 28, 2026)
Additional sourceThinking with Video, VideoThinkBench (CVPR 2026 paper) (retrieved October 19, 2026)
Additional sourceSlopTV (retrieved October 22, 2026; October 28, 2026)
Additional sourceWFM-Eval (retrieved October 23, 2026)
Additional sourceMemeBench Video Arena (retrieved October 28, 2026)
Additional sourceRapidata I2V Physics benchmark (retrieved October 22, 2026; October 28, 2026)
Additional sourceRapidata I2V Realism benchmark (retrieved October 22, 2026; October 28, 2026)
Additional sourceRapidata Relative Camera Movement (retrieved October 28, 2026)
Additional sourceModel metadata (retrieved October 16, 2026)123
Additional sourceModel metadata (retrieved October 16, 2026; October 28, 2026)
Additional sourceModel metadata (retrieved October 17, 2026)123
Additional sourceModel metadata (retrieved October 18, 2026)
Additional sourceModel metadata (retrieved October 28, 2026)123
Additional sourceGemini API Veo guide (retrieved November 5, 2026)
Additional sourceGoogle Gemini API deprecations · Editorial description reviewed October 2, 2026
Additional sourceOpenRouter: Popularity index (derived), CC BY 4.0
Additional sourceVercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional sourceOllama: cumulative downloads for the popularity index
Additional sourceReplicate: cumulative runs for the popularity index
Additional sourceGoogle Trending Now: limited feed selection for Germany and the US
Additional sourceCreative Commons Attribution 4.0