On this page
Quick answer: Test candidates with your own prompt and material. Compare inputs, controls, output settings, usable results, export and plan terms. Roundups help surface candidates; they do not replace vendor documentation or a real test.
Separate the generation tasks
| Task | What you provide | What to verify before comparing |
|---|---|---|
| Text to video | A description of scene, subject and action | Prompt controls, duration and aspect-ratio options |
| Image to video | A still image and motion description | Image input requirements and available motion controls |
| Reference-based generation | Video, image or audio references, depending on product | Whether that reference type is supported and its limits |
| Extend existing footage | An existing video clip | Whether an extension mode is explicitly provided and how it differs from generation |
| Plan multi-shot content | A script, shot list or segmented tasks | Whether the product documents this feature or requires separate generation/editing |
A product may support several tasks or only one. Do not infer image references, native audio or extension from the presence of a video-generation entry.
Build a shortlist by task
These are candidates to verify, not a quality ranking; some names are products or platforms rather than single models:
| Candidate product/series | Official information to check |
|---|---|
| Veo 3.1 (Gemini API) | Current inputs, access method, version and plan terms |
| Kling AI video tools | Current modes, inputs, output settings and regions |
| Runway platform | Current generation/editing modes, plans and export limits |
| Luma Ray | Current product name, supported inputs and available settings |
| Pika Video Studio | Current product entry, generation modes and plan limits |
| Hailuo Image to Video | Current image inputs, output options and availability |
Open each candidate’s latest product and help pages to confirm it supports the exact task you want to compare. Product names, available versions and regions can change.
Compare models on six criteria
- Inputs. Does it accept text, still images or the reference material you need? What are the format and size limits?
- Controls. Can you specify aspect ratio, resolution, duration or camera direction? Check the controls documented for the actual mode.
- Consistency. Test subject, action, background and cross-shot changes with the same material and prompt; do not rely only on selected demos.
- Audio. If sound matters, check whether it is generated by the model, added separately or made externally, using current official mode documentation.
- Iteration and export. Verify regeneration, editing, output format, clip length and export limitations.
- Plans and rights. Check allowance, membership/subscription, region, content rules and publishing terms; record the date.
Record what you verify
| Field | What to record |
|---|---|
| Candidate and official links | Current product and help pages |
| Mode/version checked | Name shown on the current page |
| Inputs and controls | Inputs, duration and output options explicitly documented |
| Plan and availability | Current price/allowance, region and account conditions |
| Test result | Prompt, reference material and exported file |
| Verification date | Date the official page was last checked |
This separates current product information from your own test results and makes later updates easier.
Decide what else the production needs
Generating a clip does not finish a video. Identify editing, captions, voiceover, music and export steps from the finished piece; verify whether each product includes them on its official site. For text-to-video free-plan comparisons, see AI video generator comparisons. For still-image input, see free image-to-video tools. For extending a short existing shot, see how to extend a video clip with AI. For the full editing workflow, see how to edit a video.
Where Aixcut fits
Aixcut can generate video clips from text and, depending on the selected model, supported reference inputs. You can preview a result, revise the description and generate again; the clip can be edited in the project with captions, voiceover and other media. AI video generation requires a membership; new accounts receive 2,000 credits. Duration and settings depend on the selected model and mode. Export uses no credits and adds no watermark.
See the Aixcut AI video generator for the current entry point, or the AI video editor for editing afterwards.
Official sources
Capabilities, allowances and plans for the third-party products in this article follow their official pages. The notes below were checked against those pages in October 2026:
| Product (official page) | What the official page states |
|---|---|
| Veo / Google DeepMind | The current model on the page is Veo 3.1, presented around native audio and extended videos; access is through Gemini and Google Flow |
| Kling AI | The site leads with Kling 4.0 (audio-visual) and Kling Image 3.0 (native 4K images); the membership page lists monthly plans at Standard $6.99 first month then $8.8/month with 660 credits, Pro $25.99 then $32.56/month with 3,000 credits, Premier $135 then $181.99/month with 18,000 credits and Ultra+ $402 then $544.99/month with 57,000 credits, and states yearly billing saves 40% |
| Runway | Free plan $0 with 125 one-time credits; Standard $15/month ($12 billed yearly) with 625 credits/month, which the page puts at roughly 52 seconds of Gen-4.5 |
| Luma Dream Machine | Plus $30/month with 10,000 credits, Pro $90/month with 40,000, Ultra $300/month with 150,000; Ray3.2 counts 400 credits for 5 seconds at 1080p |
| Pika | The Video Studio page describes output from short clips up to 30-second multi-shot pieces; Extend Video lengthens an existing clip |
| Hailuo / MiniMax | The site states MiniMax H3 and Wan 3.0 are live and offers 3,000 free credits for downloading MiniMax Design; its membership prompt starts at $9.99 (35% off for a limited time) with 1080p video creation, priority VIP generation and watermark-free commercial export |
| Google Gemini | Video generation (Gemini Omni) is a paid-plan capability and the page tells non-subscribers to upgrade; it supports extending a scene and conversational edits |
| ChatGPT | The pricing page lists Free at $0/month, Go at $8/month, Plus at $20/month and Pro from $100/month with three usage tiers; it does not list video-generation capability, so capability and allowances follow OpenAI’s official product pages and help centre |
Each line comes from the vendor’s own page, checked in October 2026. Features, allowances and prices change; the vendor’s latest page always governs.