CostPerGen

Enter a workload once and see what it costs on every indexed video model and provider, normalized to 720p output seconds. The URL updates as you type, so a filled-in calculator is shareable.

AI video cost calculator

Presets:

Model · provider$/sec (720p)TotalStatus
LTX-2 · Self-hosted (rented GPU)$0.010—estimated
LTX-2 · fal.ai (19B Distilled)$0.018—derived
Seedance 1.0 · Kie.ai$0.020—estimated
Wan 2.2 · Self-hosted (rented GPU)$0.020—estimated
Wan 2.2 · Replicate$0.021—estimated
Seedance 1.0 · RunPod Hub$0.024—derived
Seedance 1.0 · fal.ai (Lite)$0.036—estimated
LTX-2 · fal.ai (19B)$0.040—derived
Pika 2.2 · fal.ai$0.040—derived
Hailuo 02 · fal.ai (Standard (768p))$0.045—derived
Kling 2.1 · fal.ai (Standard)$0.050—estimated
Luma Ray 2 · fal.ai (Ray 2 Flash)$0.050—derived
Runway Gen-4 Turbo · Runway$0.050—derived
Wan 2.2 · RunPod Hub$0.060—derived
Kling 2.5 Turbo Pro · fal.ai$0.070—derived
Kling 2.5 Turbo Pro · Replicate$0.070—estimated
Wan 2.2 · fal.ai (A14B)$0.080—derived
Hailuo 02 · fal.ai (Hailuo 2.3 Pro)$0.082—derived
Kling 2.1 · fal.ai (Pro)$0.098—derived
Luma Ray 2 · fal.ai$0.100—derived
Sora 2 · OpenAI$0.100—estimated
Sora 2 · fal.ai$0.100—derived
Veo 3.1 Fast · Kie.ai$0.100—estimated
Runway Gen-4 · Runway$0.120—derived
Veo 3.1 Fast · Google (Gemini API / Vertex)$0.150—estimated
Veo 3.1 Fast · fal.ai$0.150—derived
Veo 3.1 · Kie.ai$0.250—estimated
Kling 2.1 · fal.ai (Master)$0.280—derived
Sora 2 Pro · OpenAI$0.300—estimated
Sora 2 Pro · fal.ai$0.300—estimated
Veo 3.1 · Google (Gemini API / Vertex)$0.400—estimated
Veo 3.1 · fal.ai$0.400—derived
Veo 3.1 · Replicate$0.400—estimated

Totals use the normalized canonical rate (USD per second of 720p output). Resolution multipliers, minimum charges, and tier effects can change real invoices — check each model page.

Also: voice calculator ·full video index ·how normalization works