Try AIContentDrop

AI video model index

Spec sheets for the models covered by this site's comparisons. Every value comes from an official source (the vendor's own pages, or a leaderboard publisher for its own ranking) and carries a checked date. Anything else is shown as “No public data”.

Live — checked 2026-10-07

Kling 3.0

Kling AI

Kling 3.0 is the video model Kling AI sells today in its app and on its developer API: single takes up to 15 s[1], a native 4K mode[1], multi-shot scenes and native audio, bought as monthly credit plans or prepaid API units[2]. Its successor Kling 4.0 is in early access from 2026-09-28, with full launch stated for October 2026[3]; Kling's API plans do not cover it[2].

Appears in: Kling vs Veo 3 · Sora vs Kling

Live — checked 2026-10-07

Veo 3.1

Google

Google's current video model: 8 s[4] clips that extend to 148 s[4] at 720p, 4K output on 8-second clips[4] and native audio[5], billed per second in the Gemini API[6].

Appears in: Kling vs Veo 3 · Veo 3 vs Sora

DISCONTINUED by vendor

Sora 2

OpenAI

OpenAI's video model, discontinued in 2026: the Sora web and apps closed on 2026-04-26[7] and the Videos API on 2026-09-24[8]. Kept here only as a sourced historical record — no Sora 2 video can be generated today.

Appears in: Veo 3 vs Sora · Sora vs Kling · Seedance 2.5 vs Sora 2

Live — checked 2026-10-07

Seedance 2.5

ByteDance

ByteDance Seed's video model: up to 30 s[9] per generation, up to 50[9] image, video and audio references in a single pass, native audio and multi-shot output[9]. Text-to-video and image-to-video are both supported[9]; ByteDance publishes no per-second API price.

Appears in: Seedance 2.5 vs Sora 2

Live — checked 2026-10-07

Veo 3.1 Lite

Google

The budget tier of Google's Veo 3.1 family, in preview in the Gemini API as veo-3.1-lite-generate-preview[10]. It is billed per second with audio included by default — $0.05/s[10] at 720p, $0.08/s[10] at 1080p — and 4K output is not supported[10]. The Gemini API free tier does not include it[10].

Appears in: Veo 3.1 vs Veo 3.1 Lite

Live — checked 2026-10-07

Runway Gen-4.5

Runway

Runway's flagship text-to-video and image-to-video model[11], released on 2025-12-01[12] and included in every paid Runway plan[12]. On the Runway Dev API it is billed at $0.12/s[11] of output, with single clips of up to 10 s[13] and 720p-class output only[14]. Runway publishes no statement on whether Gen-4.5 generates native audio.

Appears in: Kling 3.0 vs Runway Gen-4.5 · Veo 3.1 vs Runway Gen-4.5

Live — checked 2026-10-07

Luma Ray3.2

Luma AI

Luma AI's current video model, released on 2026-06-09[15] and offered as an API for the first time in the Ray line[15]. Clips run up to 20 s[15] at 24fps[16], in resolutions up to 1080p[16], with up to 16 keyframes per clip for storybeat control[16]. It generates no native audio[16].

Appears in: Luma Ray3.2 vs Kling 3.0

Live — checked 2026-10-07

MiniMax H3

MiniMax

MiniMax's general-purpose multimodal model: clips of up to 15 s[17] at 2K resolution with native stereo sound[18], billed per second on the MiniMax platform API from $0.08/s[19] at the 768P tier[19]. It ranks first[20] on LMArena's image-to-video leaderboard with a score of 1495[20], and open-weight H3-Base checkpoints are published on Hugging Face[21].

Appears in: Minimax H3 vs Kling 3.0 · Minimax H3 vs Veo 3.1

Live — checked 2026-10-07

Wan 3.0

Alibaba

Alibaba's video model, generally available through Alibaba Cloud Model Studio since 2026-08-24[22]. One generation runs up to 30 s[23] with native dialogue, BGM and sound effects[24], and a single request accepts up to 20[24] multimodal reference materials[24]. API output is billed per second by resolution tier, from $0.10/s[22] at 720P in Singapore[22].

Appears in: Wan 3.0 vs Kling 3.0

Live — checked 2026-10-07

Seedance 2.0

ByteDance

ByteDance Seed's previous-generation video model, launched on 2026-02-12[25] and still live globally through Dreamina[26]. It generates multi-shot clips of up to 15 s[25] with dual-channel audio[25] and takes up to 12[26] multimodal references[26]. ByteDance's ModelArk pricing table could not be fetched, so no official per-second API price is recorded here.

Appears in: Seedance 2.5 vs Seedance 2.0

Live — checked 2026-10-07

PixVerse V6

PixVerse

PixVerse's current model, launched on 2026-03-30[27] and offered across the consumer app, the platform API and a CLI for agentic workflows[27]. It generates multi-shot short films with native audio from a single prompt[27], in clips of up to 15 s[28] at up to 1080p[28], billed in credits per second of output[29].

Appears in: PixVerse V6 vs Kling 3.0

Live — checked 2026-10-07

Vidu Q3

Shengshu

Vidu Q3 generates finished clips with dialogue, voiceover, sound effects and music baked in, at up to 16 s[30] per generation[30]. Reference-to-video accepts up to 7[31] images[31], and resolution options run to 1080p[32]. The Open Platform API bills in credits with off-peak discounts[32]; its per-second prices are not published on a fetchable page.

Appears in: Vidu Q3 vs Kling 3.0

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”.

  1. Kling VIDEO 3.0 Model User Guide — Kling AI. kling.ai/quickstart/klingai-video-3-model-user-guide. Checked 2026-10-06
  2. Kling AI developer API pricing — Kling AI. kling.ai/dev/pricing. Checked 2026-10-06
  3. Kling 4.0 — official model announcement — Kling AI. kling.ai/blog/kling-4-ai-video-model. Checked 2026-10-06
  4. Gemini API — Generate videos with Veo — Google. ai.google.dev/gemini-api/docs/veo. Checked 2026-10-06
  5. Gemini API — Video generation docs index — Google. ai.google.dev/gemini-api/docs/video. Checked 2026-10-06
  6. Gemini API — Pricing (Veo 3.1 Standard / Fast / Lite) — Google. ai.google.dev/gemini-api/docs/pricing. Checked 2026-10-06
  7. What to know about the Sora discontinuation — OpenAI Help Center. help.openai.com/en/articles/20001152-what-to-know-about-the-sora-di…. Checked 2026-10-06
  8. OpenAI API — sora-2 model page — OpenAI. developers.openai.com/api/docs/models/sora-2. Checked 2026-10-06
  9. Introducing Seedance 2.5 — One-Take Creation, Flexible Referencing — ByteDance Seed. seed.bytedance.com/en/blog/one-take-creation-flexible-referencing-i…. Checked 2026-10-06
  10. Gemini Developer API pricing (Veo 3.1 rows) — Google. ai.google.dev/gemini-api/docs/pricing. Checked 2026-10-07
  11. Runway Dev docs — Available AI Models — Runway. docs.dev.runwayml.com/guides/models. Checked 2026-10-07
  12. Runway Research — Introducing Runway Gen-4.5 — Runway. runway.com/research/introducing-runway-gen-4.5. Checked 2026-10-07
  13. Runway Dev docs — API Changelog & Updates — Runway. docs.dev.runwayml.com/api-details/api_changelog. Checked 2026-10-07
  14. Runway Dev docs — API Input Parameters — Runway. docs.dev.runwayml.com/assets/inputs. Checked 2026-10-07
  15. Luma AI — Introducing Ray3.2 (official announcement) — Luma AI. lumalabs.ai/news/introducing-ray-3-2. Checked 2026-10-07
  16. Luma AI — Ray product page (Ray3.2 specs & FAQ) — Luma AI. lumalabs.ai/ray. Checked 2026-10-07
  17. MiniMax API Docs — Video Generation guide — MiniMax. platform.minimax.io/docs/guides/video-generation. Checked 2026-10-07
  18. MiniMax — 'MiniMax H3' launch announcement — MiniMax. www.minimax.io/blog/minimax-h3. Checked 2026-10-07
  19. MiniMax API Docs — Product Pricing (overview) — MiniMax. platform.minimax.io/docs/pricing/overview. Checked 2026-10-07
  20. Arena (arena.ai) — Image-to-Video Leaderboard — arena.ai. arena.ai/leaderboard/image-to-video. Checked 2026-10-07
  21. Hugging Face — MiniMaxAI/MiniMax-H3 model card (vendor-published) — MiniMax (MiniMaxAI on Hugging Face). huggingface.co/MiniMaxAI/MiniMax-H3. Checked 2026-10-07
  22. Alibaba Cloud Community blog — Wan3.0 at General Availability: Capabilities, Benchmarks, Pricing, and the Workflows It Changes — Alibaba Cloud. www.alibabacloud.com/blog/wan3-0-at-general-availability-capabiliti…. Checked 2026-10-07
  23. Alibaba Cloud Model Studio — wan3.0-video model information — Alibaba Cloud. www.alibabacloud.com/help/en/model-studio/wan3-0-video. Checked 2026-10-07
  24. Alibaba Cloud Model Studio — Wan3.0 Video Generation guide — Alibaba Cloud. docs.modelstudio.console.alibabacloud.com/en/model-studio/wan3-vide…. Checked 2026-10-07
  25. ByteDance Seed blog — Seedance 2.0 Official Launch — ByteDance Seed. seed.bytedance.com/en/blog/seedance-2-0-official-launch. Checked 2026-10-07
  26. Dreamina (CapCut) — Official Seedance 2.0 tool page — Dreamina / CapCut (ByteDance). dreamina.capcut.com/tools/seedance-2-0. Checked 2026-10-07
  27. PixVerse Launches V6, Advancing AI Video Generation Across Creative and Agentic Workflows — PixVerse. pixverse.ai/en/blog/pixverse-launches-v6-advancing-ai-video-generation. Checked 2026-10-07
  28. V6 — PixVerse Platform Docs (API model reference) — PixVerse. docs.platform.pixverse.ai/v6-2056814m0. Checked 2026-10-07
  29. Pricing — PixVerse Platform Docs — PixVerse. docs.platform.pixverse.ai/pricing-796039m0. Checked 2026-10-07
  30. Vidu Q3 — AI Video Model with Native Audio (official product page) — ShengShu Technology (Vidu). www.vidu.com/vidu-q3. Checked 2026-10-07
  31. Vidu Open Platform — Vidu Q3 Reference-to-Video API reference — ShengShu Technology (Vidu). platform.vidu.com/docs/api-reference/video-models/vidu-q3/reference…. Checked 2026-10-07
  32. Vidu Open Platform — Vidu Q3 Text-to-Video API reference — ShengShu Technology (Vidu). platform.vidu.com/docs/api-reference/video-models/vidu-q3/text-to-v…. Checked 2026-10-07