Blog
AI Image & Video API Guides
Hands-on model tests, production tutorials, prompt techniques, and current API pricing for generative media teams.
Hailuo 2.3 API Pricing: $0.40 per 6-Second Clip, $0.27 on Fast
Hailuo 2.3 bills per clip: $0.40 for 6 seconds and $0.80 for 10 seconds. Hailuo 2.3 Fast is image-to-video only: $0.27 for 6 seconds and $0.46 for 10. There is no resolution setting, so only duration changes the price.
HappyHorse 1.1 API Pricing: $0.16 per Second at 720p
All three HappyHorse 1.1 endpoints cost $0.16/s at 720p and $0.21/s at 1080p. Clips run 3–15 seconds; text-to-video includes audio at no extra charge. 1.1 is cheaper than 1.0: $0.21 vs $0.288 per second at 1080p.
Kling 3.0 Prompts: 6 Video Examples With Full Prompts and Prices
Split multi-shot prompts with Shot 1 / Shot 2; five seconds fit three shots. For a talking character, turn on sound and put the line in quotes. Don't ask for camera motion and a big lighting change in the same five-second clip.
Wan 2.7 API Guide: Text-to-Video and Image-to-Video in Python
Wan 2.7 runs on HiAPI as wan2.7-video/text-to-video and wan2.7-video/image-to-video, 2–15 seconds per clip. Both cost $0.10 per second at 720P and $0.167 at 1080P. Image-to-video takes a media array: first frame, last frame, driving audio, or a first clip.
Grok Imagine 1.5 API Pricing: $0.0114 per Second at 480p
Grok Imagine 1.5 public policy rates: $0.0114/s at 480p, $0.0214/s at 720p, and $0.057142/s at 1080p. A 6-second 480p clip costs $0.0684, and duration runs from 1 to 15 seconds. Previous-generation Grok Imagine text-to-video and image-to-video cost the same.
Dreamina Alternative: Image & Video Generation APIs on hiapi
Dreamina (ByteDance's Jimeng) is a consumer web/app tool with no widely available official public API — fine for hand-made clips, a dead end for backend or batch work. hiapi gives you a real image + video generation API: Nano Banana 2, Seedream 5.0 Pro and FLUX.2 for stills; Veo 3.1, Kling 3.0 Omni and Seedance 2.5 for video. It even includes the same ByteDance engines Dreamina runs on — Seedream 5.0 Pro and Seedance 2.5 — plus alternatives, all behind one key and one async /v1/tasks contract. Pay per call, not per seat: images from $0.036, video from $0.10/s.
wan3.0-video-prime Prompts: 3 Copy-Paste Recipes With Real Clips
Wan 3.0 Prime (wan3.0-video-prime) is the fast tier of the Wan 3.0 video family on HiAPI, generating 2-30 second clips at 480P/720P/1080P with optional synchronized audio. This guide gives three complete, copy-paste prompts - plain text-to-video, text-to-video with audio, and reference-to-video - each shown with the exact request and the clip it produced. Prime rewards direction over description: one subject, one camera move, one or two real motions, a stated look, and a named audio layer when you want sound. All three clips were rendered at 480P for 5 seconds ($0.4375 each); draft cheap at 480P, then re-run keepers at 720P or 1080P with the identical prompt.
Best Together AI Alternatives in 2026: Choosing an Image and Video API
Together AI started as an LLM and open-weight inference platform, but in 2026 it added a real image and video generation stack: FLUX on an OpenAI-compatible endpoint, plus a separate native endpoint for video. Its deepest strengths are still LLM serving, fine-tuning, and on-demand GPU clusters, and a per-call generation API does not replace GPU hosting or custom training. If you mainly want managed, per-call image and video generation, HiAPI, Replicate, fal.ai, and Hugging Face each cover that need with a distinct strength: breadth, catalog, speed, and open-weight flexibility.
How to Use omnihuman-1.5 via the hiapi API: curl, Python, and a Working Request
omnihuman-1.5 lip-syncs a single portrait to an audio clip and returns an MP4, using hiapi's standard POST /v1/tasks create-and-poll flow. The input is strict: image_url and audio_url are required, prompt and resolution (720p/1080p) are optional, and there is no text-only or duration field.
Seedance 2.5 API Pricing: $0.204 per Second at 720p
Text-to-video and image-to-video cost $0.204/s at 720p and $0.516/s at 1080p, audio included. Reference-to-video costs $0.1215/$0.2715/$0.68 per second at 480p/720p/1080p, billed on reference plus output length. For text-only clips under 15 seconds, Seedance 2.0 on the ext route is cheaper; use 2.5 for single takes up to 30 seconds.
Build with the models you just read about.
Compare image, video, and audio models, then use the same API workflow in production.