Step 1
Open Studio and pick a t2v model
Go to Studio → Video and choose a model from the text-to-video list on this page.
Text-to-video AI generation tools
Turn a written prompt into a finished clip. FairStack only lists models whose config type or supportedModes include text-to-video — image-only models like Veo 3.1 Fast never appear here.
Seedance 2.0 · Prompt: Morning dew on alpine grass, slow lateral track, soft backlight
What is AI text to video?
Text-to-video AI generates a short clip from a written prompt — subject, camera, and mood — without a start frame. On FairStack, only models with a t2v type or t2v supportedModes are listed.
Try text to video →Text to Video models
22 selectable models — filtered from config capability flags, not a hand-maintained list.
Seedance 2.0
$0.246/s (720p, text-to-video)
Seedance 2.0 is a text-to-video model on FairStack. $0.246/s (720p, text-to-video): base model cost + a flat 20% platform fee, shown before you spend.
Kling 3.0 Pro
$0.162/s
Kling 3.0 Pro from Kuaishou delivers strong motion quality (0.90), excellent camera control (0.85), and support for clips up to 15 seconds. When video quality is the top priority and cost is secondary, this is the definitive choice. The model produces smooth, naturalistic movement that avoids the common artifacts of AI video — jitter, ghosting, and unnatural physics. Camera control follows complex instructions including pan, zoom, dolly, and orbital movements with high fidelity. Up to 15 seconds of continuous video is supported, the longest duration among top-tier models. Visual quality (0.88 score) is consistently high across subjects and scenes. Compared to Veo 3.1 Quality (the only model with native audio), Kling 3.0 Pro offers longer maximum duration and lower cost per second. Against Grok Imagine T2V, it delivers noticeably superior motion quality and camera control at a higher price. Best suited for premium video production where quality is paramount, long-form clips up to 15 seconds, cinematic content requiring smooth camera work, and professional video workflows. Available on FairStack at infrastructure cost plus a 20% platform fee.
Seedance 2.0 Fast
$0.198/s (720p, text-to-video)
Seedance 2.0 Fast is a text-to-video model on FairStack. $0.198/s (720p, text-to-video): base model cost + a flat 20% platform fee, shown before you spend.
WAN 2.6 T2V
$0.084/s (720p)
WAN 2.6 T2V is Alibaba's latest text-to-video generation model, representing the most advanced entry in the WAN family. It generates 5-second video clips at 720p with notably strong artistic style and visual quality, making it particularly well-suited for stylized and creative content rather than strict photorealism. The model benefits from Alibaba's extensive research in video generation, delivering improved temporal coherence and motion quality over earlier WAN versions. It handles artistic prompts with distinctive flair, producing outputs with painterly qualities, rich color grading, and creative visual interpretations that set it apart from purely photorealistic competitors. Compared to Kling 3.0 which focuses on cinematic realism, WAN 2.6 T2V leans toward artistic expression and stylized output. The model is hosted on RunPod for reliable delivery. Best suited for artistic video content, music video concepts, stylized social media clips, and creative projects where visual artistry matters more than photorealism. Available on FairStack at infrastructure cost plus a 20% platform fee.
Seedance 2.0 Mini
$0.123/s (720p, text-to-video)
Seedance 2.0 Mini is ByteDance’s most affordable Seedance 2 model, built for high-volume video generation where cost matters more than maximum resolution. Like the rest of the Seedance 2 family it is multi-mode from a single endpoint: drive it with a text prompt, a single first frame, first+last frames for interpolation, or up to nine reference images plus video/audio references — FairStack auto-detects the mode from your input. It renders 480p or 720p clips from 4 to 15 seconds with native audio generation, and supports seven aspect ratios (including 21:9 and adaptive). Best for social-media shorts, rapid iteration, storyboard animation, and any workflow that generates a lot of video. Available on FairStack at infrastructure cost plus a 20% platform fee.
Grok Imagine 1.5 T2V
$0.168/s (720p)
Grok Imagine Video 1.5 is xAI's versioned 1.5 video model, generating video with audio directly from a prompt. Audio comes with the picture in the same request — there is no separate audio pass and no audio surcharge. Three resolution tiers (480p, 720p, 1080p) and any duration from 1 to 15 seconds, across seven aspect ratios including 3:2 and 2:3. Billing is per second of output and varies by tier, at the model's cost plus a flat 20% platform fee, shown before you spend. This is a separate entry from the older unversioned Grok Imagine video models, which remain available.
Grok Imagine T2V
$0.018/s (720p)
It represents the best value proposition in the premium video category — priced below Kling Pro or Veo while still producing high-quality output suitable for professional use. The model generates up to 10 seconds of video with good motion quality (0.84) and strong visual quality (0.85). The balanced capability profile means it handles diverse subjects and scenes without significant weaknesses in any particular area. As a newer entrant in the video generation space, it benefits from xAI's latest research while building a growing track record in production workflows. Compared to Kling 3.0 Pro, Grok T2V trades some motion smoothness and camera control precision for a lower cost per generation. Against Veo 3.1 Fast, it offers similar quality at a competitive price with up to 10-second duration. Ideal for general-purpose video generation, best-value premium video, 10-second clips, and any budget-conscious video workflow that still demands high quality. Available on FairStack at infrastructure cost plus a 20% platform fee.
H3 Max
$0.096/s (768p)
H3 Max is fal's post-trained variant of MiniMax H3, co-optimised with fal's inference stack. A 5-second 768P clip renders in under three seconds — faster than real time, and roughly 35x the throughput of the stock H3 endpoint. Audio is generated natively in the same pass as the picture, so dialogue, music and sound effects arrive already in sync; there is no separate audio step and no audio surcharge. Two resolutions (480P and 768P) and any duration from 5 to 15 seconds. Billing is per second of output, at the model's cost plus a flat 20% platform fee, shown before you spend. In fal's human-preference evaluations H3 Max ranked first on overall quality and prompt understanding against the field. The trade-off versus base MiniMax H3 is resolution ceiling: H3 Max tops out at 768P, so reach for stock H3 when you need 2K or 4K. For everything else — iterating on a shot, batching social cuts, anything where turnaround matters — Max is the faster half of the family. Available on FairStack at infrastructure cost plus a 20% platform fee.
HappyHorse 1.0 T2V
$0.288/s (1080p)
HappyHorse 1.0 T2V is a text-to-video model on FairStack. $0.288/s (1080p): base model cost + a flat 20% platform fee, shown before you spend.
HappyHorse 1.1 T2V
$0.174/s (1080p)
HappyHorse 1.1 is Alibaba's flagship video model, a strict upgrade over HappyHorse 1.0: lower price, built-in native audio with Foley, and multilingual lip-sync. The text-to-video mode renders 3-15 second clips at 720p or 1080p across nine aspect ratios (16:9 through 9:21). Best all-rounder when synchronized audio and speech matter.
Kling 3.0 Standard
$0.12/s
Kling 3.0 Standard is the cost-optimized variant of Kuaishou's video generation model, sharing the same architecture and maximum 15-second duration as the Pro variant while delivering good quality at a lower price point. It bridges the gap between budget video models and the premium Kling Pro tier. The model produces solid video with good motion quality (0.82 score), maintaining natural-looking movement and reasonable camera following. Supporting the same 15-second maximum as Pro means users can generate longer clips without the premium cost. The Kling architecture provides consistent visual coherence throughout the clip duration. Compared to Kling 3.0 Pro, the Standard variant shows a noticeable quality gap in motion smoothness, fine detail, and camera precision. It occupies the practical middle ground for users who need more than budget quality but less than premium pricing. Best suited for budget premium video work, longer clips where 15-second duration matters, good-enough quality workflows, and projects where Kling Pro's cost premium is not justified. Available on FairStack at infrastructure cost plus a 20% platform fee.
Kling 4.0
$0.171/s (720p)
Kling 4.0 is Kling AI's fourth-generation video model, released in October 2026. This key runs it as text-to-video: describe the shot and get a clip of any whole-second length from 3 to 30 seconds, at 720p or 1080p, in 16:9, 9:16, 1:1 or 21:9. The jump over Kling 3.0 is length and framing: runs up to 30 seconds (3.0 stopped at 15) and a cinematic 21:9 frame. For frame-by-frame control use Kling 4.0 Keyframes; to steer the shot with reference images, clips, characters or voices use Kling 4.0 Reference. Priced per second of output at the resolution you pick: the provider's rate plus a flat 20% platform fee, shown before you spend.
Kling 4.0 Reference
$0.171/s (720p)
Kling 4.0 Reference builds a shot from your references: up to 10 images, up to 5 video clips (2 to 30 seconds each), up to 7 named subjects built from 2 to 4 photos each, and up to 3 voice samples, 15 references in all. Mention them in the prompt as @image_1, @video_1 or by the subject's name (@Hero says "..."). Without reference clips the model generates its own audio; give a subject a voice sample and it speaks in that voice. With reference clips you can keep their original sound. Priced per second at the resolution you pick: the provider's rate plus a flat 20% platform fee, shown before you spend. With reference clips the per-second rate is lower, and the bill counts the clips' length as well as the output's (at least twice the output), exactly as the provider bills it.
MiniMax H3
$0.072/s (768p)
MiniMax H3 is the frontier video model that H3 Max is a post-trained variant of. Audio is generated in the same pass as the picture, so dialogue, music and effects arrive already in sync. Where H3 Max trades resolution for speed and stops at 768P, base H3 adds 2K and 4K tiers (both upscale a 768P base result). Four tiers in all — 480p, 768p, 2K, 4K — across six aspect ratios including 21:9, at any duration from 5 to 15 seconds. Billing is per second of output and rises with the tier, at the model's cost plus a flat 20% platform fee, shown before you spend. Pick base H3 when you want 2K or 4K delivery from one generation; pick H3 Max when turnaround matters more than resolution.
P-Video (Pruna)
$0.024/s (720p)
P-Video (Pruna) is a text-to-video model on FairStack. $0.024/s (720p): base model cost + a flat 20% platform fee, shown before you spend.
PixVerse V5.6 T2V
$0.108/s (720p, 5 s clip)
PixVerse V5.6 T2V is PixVerse's latest text-to-video model, known for producing high-quality video with a distinctive creative and artistic sensibility. The model excels at stylized content, generating video with strong visual aesthetics, rich color palettes, and compositional choices that lean toward artistic expression rather than strict photorealism. The price reflects the model's strong visual quality and its particular strength in creative and artistic content. The V5.6 architecture represents significant improvements over earlier PixVerse versions in both quality and consistency. Compared to photorealism-focused models like Kling 2.6, PixVerse V5.6 is stronger at stylized, artistic, and creative content but may not match the natural look of dedicated photorealistic generators. Best suited for creative video content, artistic video production, and visually appealing output where stylistic quality matters. Available on FairStack at infrastructure cost plus a 20% platform fee.
PixVerse V6 T2V
$0.054/s (720p)
PixVerse V6 is the current PixVerse generation. It generates short video from a text prompt across four resolution tiers (360p, 540p, 720p, 1080p) and eight aspect ratios including 21:9, at any duration from 1 to 15 seconds — V5.6 accepted only 5, 8 or 10. It also offers five stylised looks — anime, 3D animation, clay, comic and cyberpunk — selectable per generation. Billing is per second of output and rises with the tier, at the model's cost plus a flat 20% platform fee, shown before you spend. On FairStack this model renders silent video. The provider's optional soundtrack is priced on a different scale for every resolution tier, which our billing cannot represent yet, so audio is switched off rather than charged incorrectly.
Seedance 2.5
$0.378/s (720p)
Seedance 2.5 is the next generation of ByteDance's Seedance video model. Like the rest of the Seedance family it is multi-mode from one model: drive it with a text prompt, a single first frame, a first and last frame for interpolation, or a set of reference images and audio clips. FairStack detects the mode from your input. It renders 480p, 720p, or 1080p in seven aspect ratios (including 21:9 and adaptive), with synchronized audio generated in the same pass. On FairStack a clip runs from 4 to 16 seconds. The model itself supports up to 30 seconds and larger reference sets; FairStack does not expose those yet. Reference videos are not accepted on FairStack yet. Billing is per second of output at the provider rate for the resolution you choose, plus the flat 20% platform fee, shown before you spend.
Veo 3.1 Quality
$0.19125/s (1080p, 8 s clip)
Google Veo 3.1 Quality is the premium tier of Google's video generation model and the only video model that generates both 1080p video and synchronized audio in a single generation pass. No other model combines video and audio generation into one step, eliminating the need for separate audio production and synchronization. Top-tier visual quality (0.92 score) and motion quality (0.90) place it at the pinnacle of AI video generation capability. The 1080p output resolution is the highest available from any video model, delivering detail that is suitable for professional presentations and large-screen viewing. Native audio generation produces synchronized sound effects, ambient audio, and environmental sounds that match the visual content. Compared to Kling 3.0 Pro, Veo 3.1 Quality matches or exceeds visual quality while adding native audio and higher resolution. Against a workflow combining a separate video model plus audio generation, the one-step approach is simpler but may cost more. Maximum duration is 8 seconds per generation. Best suited for professional video production where both 1080p resolution and synchronized audio are required, premium marketing content, and workflows where the convenience of one-step video-plus-audio generation justifies the premium price. Available on FairStack at infrastructure cost plus a 20% platform fee.
Vidu Q3 T2V
$0.1848/s (720p)
Vidu Q3 T2V is Vidu's text-to-video model with flexible per-second billing, generating balanced-quality video from text prompts at proportional cost. The model delivers reliable output with good motion quality and scene coherence, providing a solid option for general-purpose video generation with the advantage of duration-flexible pricing. This pricing structure is advantageous for short clips where flat-rate models may overprice the output, and remains competitive for longer sequences. The model handles a variety of prompt types with consistent quality. The model provides a good balance of quality and pricing flexibility. Best suited for flexible-duration video generation and general text-to-video workflows where per-second billing provides cost control. Available on FairStack at infrastructure cost plus a 20% platform fee.
Vidu Q3 Turbo T2V
$0.0924/s (720p)
Vidu Q3 Turbo T2V is Vidu's budget text-to-video model optimized for fast generation at significantly lower cost than the standard Q3 tier. The model supports up to 16 seconds of video duration with built-in audio generation, offering one of the longest maximum durations among budget video models. Multiple resolution options are available, with 720p providing the best quality-to-cost balance. Best suited for draft video previews, social media content, and high-volume video production where budget-friendly per-second pricing enables cost-effective generation at scale. Available on FairStack at infrastructure cost plus a 20% platform fee.
WAN 3
$0.24/s (1080p)
WAN 3.0 is the current generation of Alibaba's WAN family, released in August 2026. It generates audio in the same pass as the picture, so dialogue, music and effects arrive already in sync, and it renders natively at up to 1080p. Three resolution tiers (480p, 720p, 1080p) and an adaptive aspect mode that lets the model choose framing from the prompt. Billing is per second of output, at the model's cost plus a flat 20% platform fee, shown before you spend. Upstream, WAN 3 accepts up to 30 seconds in a single generation. On FairStack it is currently capped at 16 seconds, which is the platform-wide ceiling for every video model — so the longest-form capability is not yet exposed here. Available at infrastructure cost plus a 20% platform fee.
How to use text to video
Step 2
Write a motion-first prompt
Lead with camera movement, then subject action, then lighting. Keep one idea per shot.
Step 3
Set duration and resolution
Pick length and resolution the model supports. Price scales per second.
Step 4
Generate and refine
Run the job, review the MP4, then iterate the prompt or switch models without changing accounts.
Step 5
Export or chain
Download the clip or feed it into image-to-video / video-to-video for a second pass.
Tips for text to video
Lead with camera language
Start with dolly, pan, orbit, or static locked-off — models latch onto the first motion cue.
One subject per shot
Crowded prompts smear identity. Name one hero subject and one secondary.
Name the lens
35mm, 85mm, anamorphic — focal length cues stabilize framing.
Budget the seconds
Price is per second. Draft at shorter durations, then spend on the keeper.
What creators build with text to video
Social b-roll on demand
Prompt atmosphere plates instead of licensing stock for every cut.
Concept pitches
Show a moving board before you commit to a shoot day.
Product mood films
Generate brand-safe mood clips, then composite real product plates.
Frequently asked questions
Which models support text-to-video? +
Every model on this page is filtered from config: type t2v, or supportedModes including t2v. The list updates when config changes — it is not hand-maintained.
Why isn't Veo 3.1 Fast here? +
Veo 3.1 Fast on FairStack is image-to-video only (type i2v). It appears on /video/image-to-video and is structurally excluded from this page.
How much does text-to-video cost? +
Per second at each model's charged rate. Example: Kling 3.0 Standard is $0.12/sec; Seedance 2 at 720p t2v is $0.246/sec. No invented clip pricing.
Still have questions? We're here to help.