Quick verdict
Veo 3.1 wins the head-to-head evidence. The one leaderboard carrying both, LMArena text-to-video, places veo-3.1-audio at rank #12[8] with 1364[8] and runway-gen-4.5 at rank #27[9] with 1225[9], and Veo's documented ceiling — 4K with native audio[1] — sits far above Gen-4.5's 1280:720 maximum[7]. Runway Gen-4.5 is the budget route to a current flagship: the full model ships on a $15.00/mo[10] plan[10], while full Veo 3.1 access needs Google AI Ultra from $99.99/mo[11][11].
- Pricing Tie
- Output specs Tie
- Features Tie
- Access Tie
- Public leaderboards Veo 3.1
Category winners are computed from the sourced rows in the tables below; “Depends” means there was not enough comparable public data to call it.
Summary
Runway shipped Gen-4.5 on 2025-12-01[5] and calls it its state-of-the-art text- and image-to-video model[4], available in its web app, mobile apps and Dev API[5]. Google's Veo 3.1, in the Gemini API since 2025-10-15[12] as a paid preview[12], also runs in Vertex AI, the Gemini app and Flow. Unusually for this market, one public board scores them against each other directly.
On output, the two aren't close. Veo 3.1 documents 720p, 1080p and 4K (the higher tiers on 8-second clips), 24fps, extension to 148 s[1], 3[1] reference images and first-and-last-frame transitions[1], all with native audio[2]. Gen-4.5 documents 2-10 second durations[6] and 720p-class output in seven aspect ratios[7]; audio, extension and reference input go unmentioned in Runway's documentation, so those comparisons have no public data.
Money is where Runway argues back. Its API charges a flat $0.12/s[4][4] — above Veo 3.1 Fast at $0.10/s[3] per 720p second but well under Standard at $0.40/s[3][3]. On subscriptions, Gen-4.5 comes with every paid Runway plan from $15.00/mo[10][10], whereas Google AI Pro at $19.99/mo[11] includes only a limited Veo 3.1 Lite trial[11]. For a flat-credit alternative, AIContentDrop's plan tiers bundle Veo 3.1 from $19/mo[13].
Snapshot
| Spec | Veo 3.1 | Runway Gen-4.5 |
|---|---|---|
| Status | Paid preview in the Gemini API (Google's current model list names veo-3.1-generate-preview and veo-3.1-lite-generate-preview as Preview models; Veo 3.1 Fast rates are still published on the pricing page); also served via Vertex AI, the Gemini app and Flow[12] | Generally available (all paid Runway consumer plans; API via Runway Dev since 2026-02-10)[5] |
| Developer | Google[12] | Runway[4] |
| Released | 2025-10-15[12] | 2025-12-01[5] |
| Lowest API price (per second, with audio) | from $0.05/s (Lite)[3] | from $0.12/s (gen4.5 (Runway Dev API))[4] |
| Cheapest paid consumer plan | $19.99/mo Google AI Pro $99.99/mo Google AI Ultra[11] | $15.00/mo (Standard)[10] |
| Free tier | Gemini API free tier: Not available for Veo 3.1 Standard, Fast or Lite (paid tier only)[3] | Free plan: one-time deposit of 125 credits that doesn't expire; Gen-4.5 itself is only on paid plans[10] |
| Max length, single generation | 8 s[1] | 10 s (gen4.5 (Runway Dev API))[6] |
| Max resolution | 4k (8s length only)[1] | Landscape 1280:720 and Portrait 720:1280 outputs only (text-to-video); image-to-video adds 1584:672, 1104:832, 832:1104, 672:1584 and square 960:960[7] |
| Native audio | Yes[2] | No public data |
| Image-to-video | Yes[2] | Yes[4] |
| Public API | Yes[12] | Yes (Runway Dev)[6] |
Pricing
Tie| Metric | Veo 3.1 | Runway Gen-4.5 | Leader |
|---|---|---|---|
| API price, 720p with audio (per second) | $0.40/s Standard $0.10/s Fast $0.05/s Lite[3] | $0.12/s (gen4.5 (Runway Dev API))[4] | Veo 3.1 |
| API price, 1080p with audio (per second) | $0.40/s Standard $0.12/s Fast $0.08/s Lite[3] | No public data | — |
| API price, 4K (per second) | $0.60/s Standard $0.30/s Fast[3] | No public data | — |
| API billing basis | Per second of successfully generated video, with audio included by default; price varies by tier (Standard / Fast / Lite) and by output resolution (720p / 1080p / 4k)[3] | Per second of output, flat per model: gen4.5 = 12 credits/second; credits purchased at $0.01 per credit in the developer portal (= $0.12/sec as shown on the models page); no resolution or audio price tiers for Gen-4.5[14] | — |
| Cheapest paid consumer plan | $19.99/mo Google AI Pro $99.99/mo Google AI Ultra[11] | $15.00/mo (Standard)[10] | Runway Gen-4.5 |
| Consumer plan ladder | Google AI Pro $19.99/month (1,000 Google Flow Credits, limited trial of Veo 3.1 Lite); Google AI Ultra from $99.99/month (10,000 or 25,000 Google Flow Credits, full Veo 3.1)[11] | Free $0 (125 one-time credits); Standard $15/mo monthly or $12/mo annual, 625 credits/mo; Pro $35/mo monthly or $28/mo annual, 2250 credits/mo; Max $95/mo monthly or $76/mo annual, 9500 credits/mo (early access to new models). Gen-4.5 included in all paid plans.[10] | — |
| In-app cost per second | No public data | No public data | — |
| Free tier | Gemini API free tier: Not available for Veo 3.1 Standard, Fast or Lite (paid tier only)[3] | Free plan: one-time deposit of 125 credits that doesn't expire; Gen-4.5 itself is only on paid plans[10] | — |
Veo 3.1 leads 1 row, Runway Gen-4.5 leads 1, 0 tied; rows marked — have no comparable data or no objective “better”.
Output specs
Tie| Metric | Veo 3.1 | Runway Gen-4.5 | Leader |
|---|---|---|---|
| Max length, single generation | 8 s[1] | 10 s (gen4.5 (Runway Dev API))[6] | Runway Gen-4.5 |
| Max length with extensions | 148 s (with extensions, 720p)[1] | No public data | — |
| Max output resolution | 2160p (4K) (Standard)[1] | 720p (text-to-video)[7] | Veo 3.1 |
| Frame rate | 24 fps[1] | No public data | — |
| Aspect ratios | 16:9 (default), 9:16[1] | Text-to-video: 1280:720, 720:1280. Image-to-video: 1280:720, 1584:672, 1104:832, 720:1280, 832:1104, 672:1584, 960:960[7] | — |
| Max reference inputs per generation | 3[1] | No public data | — |
Veo 3.1 leads 1 row, Runway Gen-4.5 leads 1, 0 tied; rows marked — have no comparable data or no objective “better”.
Features
Tie| Metric | Veo 3.1 | Runway Gen-4.5 | Leader |
|---|---|---|---|
| Native audio | Yes[2] | No public data | — |
| Lip-sync | No public data | No public data | — |
| Text-to-video | Yes[2] | Yes[4] | Tie |
| Image-to-video | Yes[2] | Yes[4] | Tie |
| First / last frame control | Yes[12] | No public data | — |
| Multi-shot in one generation | No public data | No public data | — |
| Video editing | Yes (Lite)[15] | No public data | — |
| Video extension | Yes[2] | No public data | — |
Veo 3.1 leads 0 rows, Runway Gen-4.5 leads 0, 2 tied; rows marked — have no comparable data or no objective “better”.
Access
Tie| Metric | Veo 3.1 | Runway Gen-4.5 | Leader |
|---|---|---|---|
| Status | Paid preview in the Gemini API (Google's current model list names veo-3.1-generate-preview and veo-3.1-lite-generate-preview as Preview models; Veo 3.1 Fast rates are still published on the pricing page); also served via Vertex AI, the Gemini app and Flow[12] | Generally available (all paid Runway consumer plans; API via Runway Dev since 2026-02-10)[5] | — |
| Public API | Yes[12] | Yes (Runway Dev)[6] | Tie |
| Where to use it | Gemini API in Google AI Studio and Vertex AI; Gemini app; Flow[12] | Runway web app (app.runwayml.com), Runway iOS/Android mobile apps, Runway Dev API (docs.dev.runwayml.com)[5] | — |
| Regional limits | In EU, UK, CH and MENA locations, allow_adult is the only permitted personGeneration value[1] | No public data | — |
| Release date | 2025-10-15[12] | 2025-12-01[5] | — |
Veo 3.1 leads 0 rows, Runway Gen-4.5 leads 0, 1 tied; rows marked — have no comparable data or no objective “better”.
Public leaderboards
Winner: Veo 3.1| Metric | Veo 3.1 | Runway Gen-4.5 | Leader |
|---|---|---|---|
| LMArena text-to-video score | 1364 ±14 (veo-3.1-audio)[8] | 1225 ±8 (runway-gen-4.5)[9] | Veo 3.1 |
| LMArena text-to-video rank (of 48) | #12 (veo-3.1-audio)[8] | #27 (runway-gen-4.5)[9] | — |
| LMArena image-to-video score | 1398 ±10 (veo-3.1-audio)[16] | No public data | — |
| LMArena image-to-video rank (of 48) | #13 (veo-3.1-audio)[16] | No public data | — |
| VGI-Bench overall score | 32.0[17] | No public data | — |
Veo 3.1 leads 1 row, Runway Gen-4.5 leads 0, 0 tied; rows marked — have no comparable data or no objective “better”.
How rows are judged: a lower price, a longer clip, a higher resolution or frame rate, more reference inputs and a supported feature win. Prices of a discontinued product are shown as historical and cannot win. LMArena scores are human-preference ratings (board snapshot dated 2026-09-21, read 2026-10-07); scores whose ± intervals overlap count as a tie. VGI-Bench is a research benchmark of visual reasoning. A row marked — has no comparable data.
Where each model leads
Where Veo 3.1 leads
Spec ceiling and scoreboard. The higher arena score, the only 4K option in this pair[1], native audio on every clip, and extensions that reach 148 s[1] where Runway documents no extension feature at all. You can also generate Veo 3.1 clips through AIContentDrop without a Google plan.
Where Runway Gen-4.5 leads
Cost of entry. One price per second on the API, 12 credits at $0.01 each[14], and the full model on the $15.00/mo[10] Standard plan — or $12 per month billed annually — with 625 monthly credits[10]. Getting unrestricted Veo 3.1 as a consumer means Google AI Ultra from $99.99/mo[11][11], more than six times Runway's entry price.
Which should you choose?
Frequently asked questions
Which ranks higher where both are listed?
Veo 3.1. On LMArena text-to-video, veo-3.1-audio holds rank #12[8] with 1364[8] against rank #27[9] with 1225[9] for runway-gen-4.5 — a gap well beyond the confidence intervals. On image-to-video, only Veo is listed.
Does Gen-4.5 generate audio?
Runway doesn't say. Its model, pricing and input-parameter pages carry no audio statement for Gen-4.5 either way, while Google documents Veo 3.1 as generating video with native audio[2]. On the published record, Veo is the only one of the two with sound.
Which is cheaper to run through the API?
It depends on the tier you'd otherwise pick. Runway charges a flat $0.12/s[4][4]. Veo 3.1 spans $0.05/s[3] (Lite, 720p) to $0.60/s[3] (Standard, 4K)[3] — so Veo Lite and Fast undercut Runway at 720p, while Veo Standard costs more than triple Runway's rate.
Evidence
Only official sources are cited: vendor pages, and leaderboard publishers for their own rankings. Anything we could not find on such a page is shown as “No public data”.
- Gemini API — Generate videos with Veo — Google. ai.google.dev/gemini-api/docs/veo. Checked 2026-10-06
- Gemini API — Video generation docs index — Google. ai.google.dev/gemini-api/docs/video. Checked 2026-10-06
- Gemini API — Pricing (Veo 3.1 Standard / Fast / Lite) — Google. ai.google.dev/gemini-api/docs/pricing. Checked 2026-10-06
- Runway Dev docs — Available AI Models — Runway. docs.dev.runwayml.com/guides/models. Checked 2026-10-07
- Runway Research — Introducing Runway Gen-4.5 — Runway. runway.com/research/introducing-runway-gen-4.5. Checked 2026-10-07
- Runway Dev docs — API Changelog & Updates — Runway. docs.dev.runwayml.com/api-details/api_changelog. Checked 2026-10-07
- Runway Dev docs — API Input Parameters — Runway. docs.dev.runwayml.com/assets/inputs. Checked 2026-10-07
- Text-to-Video Arena leaderboard — LMArena (arena.ai). arena.ai/leaderboard/text-to-video. Checked 2026-10-06
- Arena (arena.ai) — Text-to-Video Leaderboard — LMArena / arena.ai. arena.ai/leaderboard/text-to-video. Checked 2026-10-07
- Runway — Pricing (consumer plans) — Runway. runway.com/pricing. Checked 2026-10-07
- Google AI plans (Google One, US pricing) — Google. one.google.com/intl/en_us/about/google-ai-plans. Checked 2026-10-06
- Introducing Veo 3.1 and new creative capabilities in the Gemini API — Google Developers Blog. developers.googleblog.com/introducing-veo-3-1-and-new-creative-capa…. Checked 2026-10-06
- AIContentDrop — Pricing — AIContentDrop. aicontentdrop.com/pricing. Checked 2026-10-07
- Runway Dev docs — API Pricing & Costs — Runway. docs.dev.runwayml.com/guides/pricing. Checked 2026-10-07
- Gemini API — Models list (Veo 3.1, Veo 3.1 Lite) — Google. ai.google.dev/gemini-api/docs/models. Checked 2026-10-06
- Image-to-Video Arena leaderboard — LMArena (arena.ai). arena.ai/leaderboard/image-to-video. Checked 2026-10-06
- VGI-Bench leaderboard (EMNLP 2026) — VGI-Bench research team. hexuan21.github.io/VGI-Bench. Checked 2026-10-06