Media pricing
fal.ai image, video, and audio pricing per output or generation
Example catalog rates: Recraft 20b $0.0219 per image, Veo3.1 Lite Text to Video $0.05 per second, Elevenlabs Tts Eleven V3 $0.1 per 1,000 characters. fal.ai image, video, and audio pricing per output or generation is built for product teams that need the real fal.ai price per output or generation for image, video, audio, and 3D workloads. Use it to decide which fal.ai model delivers the required output at a sustainable per-run and per-accepted-output cost. Keep the workload assumptions consistent across options, then inspect the cited prices and last-checked dates before committing budget.
Open fal.ai pricing - See the real cost of each generation →
Example fal.ai prices by output
Eight real catalog rates from the ByteCosts fal.ai snapshot (1,515 models, catalog snapshot Oct 9, 2026). Each model bills per its own unit, so a per-megapixel rate scales with resolution and a per-second rate with clip length.
| Model | Price | Billing unit | Good for |
|---|---|---|---|
| FLUX.1 [schnell] (fal-ai/flux/schnell) | $0.003 | per megapixels | Draft image, billed by resolution |
| FLUX.1 [dev] (fal-ai/flux/dev) | $0.025 | per megapixels | Quality image, billed by resolution |
| Recraft 20b (fal-ai/recraft-20b) | $0.0219 | per images | Image, flat price per picture |
| Ideogram V2A Turbo (fal-ai/ideogram/v2a/turbo) | $0.025 | per images | Fast image with text, per picture |
| MiniMax (Hailuo AI) Text to Image (fal-ai/minimax/image-01) | $0.01 | per images | Budget image, per picture |
| Veo3.1 Lite Text to Video (fal-ai/veo3.1/lite) | $0.05 | per seconds | Short video clip, billed by length |
| Kling Video v3 Text to Video [Standard] (fal-ai/kling-video/v3/standard/text-to-video) | $0.14 | per seconds | Short video clip, billed by length |
| Elevenlabs Tts Eleven V3 (fal-ai/elevenlabs/tts/eleven-v3) | $0.1 | per 1000 characters | Voiceover, billed by script length |
Worked example: cost per finished output
Take Recraft 20b at $0.0219 per image: needing 3 tries for one image you can actually use costs 3 × $0.0219 = $0.0657, about $0.07 per finished image. For video, an 8-second Veo 3.1 Lite clip at $0.05 per second costs 8 × $0.05 = $0.40 per try, so 2 tries for one usable clip is $0.80. Per-megapixel prices scale with resolution and per-second prices with clip length, so compare models only after converting to the same finished output.
A lower price per try can still lose per finished output when it needs more retries: divide the cost per try by your measured acceptance rate before choosing.
The decision this page helps you make
Calculate fal.ai image, video, audio, and 3D model cost per generation from source-backed per-output prices, then compare the models side by side.
The practical question is which fal.ai model delivers the required output at a sustainable per-run and per-accepted-output cost. Use the same workload assumptions for every option so the comparison reflects billing differences instead of different inputs.
Start with these inputs
- Units: Image, megapixel, second, or generation.
- Models: Image, video, audio, 3D generation.
- Output: Per-run and accepted-output cost.
What the result includes
| Area | What ByteCosts shows |
|---|---|
| Units | Image, megapixel, second, or generation |
| Models | Image, video, audio, 3D generation |
| Output | Per-run and accepted-output cost |
How to use the result
- Run a realistic base case and a heavier-usage case before choosing a provider or plan.
- Compare alternatives with identical traffic, token, seat, runtime, and retry assumptions.
- Open the cited provider source before a purchase or production billing decision.
Formula
acceptedOutputCost = listedUnitPrice * billedUnitsPerAttempt * attemptsPerAcceptedOutput. The billed unit may be an image, megapixel, video second, audio second, or another provider-defined unit.
Assumptions
- Visible prices come from committed fal.ai pricing records and preserve the provider’s billing unit.
- Retries and rejected outputs are user-entered product assumptions, not provider facts.
- Resolution, duration, model version, and optional features can change the billed unit.
- The estimate excludes taxes, negotiated rates, and downstream storage or delivery cost.
Example scenario
Choose a model and its real billing unit, then multiply the listed per-attempt price by the number of attempts normally required to get one usable output. For video, include the requested duration when the provider bills per second.
How to read the example
| Step | Example input | What to inspect |
|---|---|---|
| Provider unit | Image, megapixel, second, or generation | What the rate actually bills |
| One attempt | Unit price times billed units | Raw generation cost |
| Accepted output | One-attempt cost times retry count | Product-level cost per usable result |
Interpretation guide
- Do not compare two models until their billing units have been normalized for the same output.
- A cheaper attempt can still be more expensive per accepted output when retries are high.
- Recheck the source record before publishing customer-facing generation prices.
Limitations
fal.ai image, video, and audio pricing per output or generation is a planning tool, not a billing guarantee. It uses the visible assumptions and committed source-backed data available at the page’s last update.
Check the cited provider page and your own production logs before signing a contract, changing price, or committing infrastructure spend.
Frequently asked questions
What should I enter first in fal.ai image, video, and audio pricing per output or generation?
Start with units: image, megapixel, second, or generation. Add optional adjustments only after the base case is understandable.
Is the result a guaranteed invoice forecast?
No. It is a planning estimate based on the visible workload assumptions and source-backed public prices. Taxes, negotiated discounts, undocumented limits, and production behavior can change the final invoice.
Where do the prices and assumptions come from?
ByteCosts keeps provider source links, confidence information, and last-checked dates attached to pricing records. User-entered workload assumptions remain separate from published vendor facts.
fal.ai image, video, and audio pricing per output or generation. ByteCosts. https://bytecosts.com/tools/fal-pricing/