Blog
AI Image & Video API Guides
Hands-on model tests, production tutorials, prompt techniques, and current API pricing for generative media teams.
Kling 2.5 Turbo API Pricing: $0.06 per Second, $0.30 per 5-Second Clip
Kling 2.5 Turbo text-to-video and image-to-video both cost $0.06 per second: $0.30 for 5 seconds, $0.60 for 10. Only length affects price; aspect ratio and first or last frames are free. Text-to-video includes audio. The same 10-second 1080p clip with sound is $1.60 on Kling 3.0 Turbo.
Kling 2.5 Turbo vs Kling 3.0 Turbo: Which Should You Use?
A 10-second 1080p clip with sound is $0.60 on Kling 2.5 Turbo and $1.60 on Kling 3.0 Turbo. 3.0 Turbo adds 3–15 second lengths and a 720p tier; 2.5 Turbo adds a last frame, six image-to-video ratios and a negative prompt. Audio differs on image-to-video: silent on 2.5 Turbo, included on 3.0 Turbo.
How to Use the Kling 2.5 Turbo API: First Video in 5 Minutes
Two endpoints, kling-2.5-turbo/text-to-video and kling-2.5-turbo/image-to-video, both $0.06 per second. Text-to-video always includes audio; image-to-video needs a first frame, takes an optional last frame and offers six ratios. cfg_scale runs 0–1; there's no seed, resolution or audio switch.
Runway Gen-4.5 API Pricing: $0.1715 per Second at 720p
Runway Gen-4.5 text-to-video and image-to-video both cost $0.1715 per second: $0.8575 for 5 s, $1.372 for 8 s, $1.715 for 10 s. Output is fixed at 720p with no audio, so only length moves the price. Vertical 9:16 text-to-video has no 8-second option.
Runway Gen-4.5 vs Kling 2.5 Turbo: Which Should You Use?
A 10-second clip is $1.715 on Runway Gen-4.5 and $0.60 on Kling 2.5 Turbo. Runway adds an 8-second option; Kling adds 1080p, a last frame, 21:9 and audio on text-to-video. Both share one API key and the same task endpoint.
Kling 2.5 Turbo Prompts: 6 Ready-to-Use Templates with Real Outputs
Six scenes, each with a real Kling 2.5 Turbo output and a full prompt to copy. Each clip is $0.30 (5 s) or $0.60 (10 s); the links pre-fill the prompt. For image-to-video write only motion and camera; for text-to-video, one paragraph covering subject, action and camera.
How to Use the Runway Gen-4.5 API: First Video in 5 Minutes
Two endpoints, runway-gen-4.5/text-to-video and runway-gen-4.5/image-to-video, both $0.1715 per second. 5, 8 or 10 seconds, fixed 720p, no audio; 9:16 text-to-video has no 8-second option. Image-to-video takes exactly one first frame and offers five ratios.
Runway Gen-4.5 Prompts: 6 Ready-to-Use Templates with Real Outputs
Six scenes, each with a real Runway Gen-4.5 output and a full prompt to copy. Each clip is $0.8575 (5 s), $1.372 (8 s) or $1.715 (10 s); the links pre-fill the prompt. One action and one camera move per prompt.
DeepSeek V4.1-Flash-1017: API Pricing, Benchmarks, Access
DeepSeek V4.1-Flash-1017 is live on HiAPI: same model ID (`deepseek-v4.1-flash`), same call pattern, zero migration work. Coding is the headline. DeepSWE v1.1 jumps from 74.2 to 77.0, and the model ranks #1 among open-source models on the Code Arena web-dev leaderboard, ahead of Kimi K3. On HiAPI it costs $0.44 per 1M input tokens and $1.32 per 1M output tokens, with cache-hit input at $0.014 — about 5x cheaper on input than Kimi K3. You get a 1M-token context, a platform-verified 393K max output, image input, and both Responses and Chat Completions endpoints with streaming. All prices in this post are as of Oct 2026.
Vidu Q4 Preview Image-to-Video: Turn a Product Still Into a 4K Short-Form Clip
vidu-q4-preview/image-to-video animates one first-frame still into a 3-16s clip up to 4K with a native audio track -- the end-to-end short-form pipeline, not a prompt list. Built on a real 2026-10-19 run: three vertical beats (4K hook, 1080p body, 720p CTA) produced, batched async, mastered and exported to H.264, with the exact schema, costs, and gotchas.
Build with the models you just read about.
Compare image, video, and audio models, then use the same API workflow in production.