NewPoint any agent at UberLLM — one key, every model, list price or less.
UberLLM

Cheapest Veo 3 API (October 2026)

Updated October 31, 2026, 17:00 UTC. UberLLM prices are read live when the page is built (every hour); other providers' prices were checked October 30, 2026 or read from their public price lists at build. UberLLM wrote this page and is one of the providers in it.

On October 31, 2026, the lowest price we found for an 8-second Veo 3.1 Fast clip at 720p with sound was APIMart’s “Ext” channel at $0.14 (sound not stated). UberLLM charges $1.02 for that clip ($0.1275 per second), 4th of 7: above APIMart and kie.ai, below fal, Replicate and WaveSpeedAI. Google shut down Veo 3.1 Preview on the Gemini API on October 22, 2026. On UberLLM a video that fails costs nothing, and the reply states the most a job can cost before it runs.

ProviderVeo 3.1 Fast, 720p with sound, 8 sFailed videos
APIMart$0.14“Failed Veo 3.1 tasks are not charged” (its page)
kie.ai$0.30“Failed tasks are never charged” (its page)
UberLLM$1.02Not charged
fal$1.20Server errors are not charged; content refusals (HTTP 422) may be
Replicate$1.20“If a run fails, we don’t charge you” (its billing docs)
WaveSpeedAI$1.20
Google Vertex AI$0.10 “/ 1 count”

Ordered by the price of an 8-second Veo 3.1 Fast clip at 720p with sound, lowest first. Google Vertex AI is last because its price page lists Veo per “count” without defining it, so it cannot be ranked. An empty cell means we could not confirm it on the provider's own pages.

Veo 3.1 Fast, 720p with sound, 8-second clip

ProviderPer 8-second clipPrice as listedNote
APIMart, “Ext” channel$0.14$0.14 per video, billed per 8-second videosound not stated; APIMart says the official channel is for 4 or 6-second clips, an audio switch or a negative prompt
kie.ai$0.30per video, the same price for 4, 6 or 8 seconds
APIMart, official channel$0.64$0.08 per second (720p with audio)
UberLLM$1.02$0.1275 per second
fal$1.20$0.15 per second
Replicate$1.20$0.15 per second
WaveSpeedAI$1.20$0.15 per second
Google Vertex AI$0.10 “/ 1 count”Google’s page does not define a count, so it is not ranked; the GA model is not sold pay-as-you-go (fixed quota or Provisioned Throughput only)

UberLLM: its price per second now, read October 31, 2026; the 5.5% card fee is not in the cell. kie.ai: its public price list, read when this page was built. fal: Fal’s own per-second price as our catalog lists it (matches fal.ai on 2026-10-30). Everyone else: their own pages, checked 2026-10-30. Per-second prices are multiplied by 8.

Veo 3.1 (Quality), 720p with sound, 8-second clip

ProviderPer 8-second clipPrice as listedNote
APIMart, “Ext” channel$1.00$1.00 per video, billed per 8-second videosound not stated
kie.ai$1.25per video, the same price for 4, 6 or 8 seconds
APIMart, official channel$2.56$0.32 per second (720p with audio)
01Router$2.88$0.36 per second (720p/1080p with audio)
UberLLM$3.17$0.396 per secondone setting: no resolution or sound option
fal$3.20$0.40 per second
Replicate$3.20$0.40 per second
WaveSpeedAI$3.20$0.40 per second
Google Vertex AI$0.40 “/ 1 count”not ranked: the count is not defined

Same sources as above. UberLLM’s Veo 3.1 has one price for every request: no resolution or sound option.

Veo 3.1 Lite, 720p, 8-second clip

ProviderPer 8-second clipPrice as listedNote
APIMart, “Ext” channel$0.07$0.07 per video, billed per 8-second videosound not stated
kie.ai$0.15per video, the same price for 4, 6 or 8 seconds
Google Vertex AI$0.05 “/ 1 count” (720p with audio)not ranked: the count is not defined; Lite is in Preview on Vertex

UberLLM does not sell Veo 3.1 Lite yet. Same sources as above.

Each provider, good and bad

APIMart

Pick it if: The lowest prices listed for Veo 3.1, if a fixed 8-second clip works for you.

Watch for: Its cheapest prices are on what it calls the “Ext” channel; sound is not stated there.

Source: APIMart price (checked 2026-10-30)

kie.ai

Pick it if: The lowest per-video Veo prices with an official price per clip, and credits that never expire.

Watch for: Each model has its own API path rather than one OpenAI-style endpoint, and there is no official n8n node or MCP server.

Source: kie.ai price (checked 2026-10-30)

UberLLM

Pick it if: One key for Veo, Kling, Seedance and every chat model, the cost known before the job starts, and no charge for failed videos.

Watch for: Not the lowest Veo Fast price: kie.ai and APIMart sell an 8-second clip for less. No Veo 3.1 Lite yet.

Source: UberLLM price (live, October 31, 2026)

fal

Pick it if: Day-one access to new video models, an official n8n node and an MCP server.

Watch for: Fal’s own Veo price is above several resellers.

Source: fal price (checked 2026-10-30)

Replicate

Pick it if: Many models with versioned, stable endpoints.

Watch for: Same Veo price as fal.

Source: Replicate price (checked 2026-10-30)

WaveSpeedAI

Pick it if: Official n8n and ComfyUI nodes, and an MCP server.

Watch for: Same Veo price as fal and Replicate.

Source: WaveSpeedAI price (checked 2026-10-30)

Google Vertex AI

Pick it if: Google’s own GA Veo 3.1 with enterprise terms.

Watch for: The GA models are not sold pay-as-you-go: fixed quota or Provisioned Throughput only. Prices are listed per “count”.

Source: Google Vertex AI price (checked 2026-10-30)

How to switch to UberLLM

One request starts a Veo job on UberLLM; check on it, then download the file. The price is known before the job starts.

bash
curl https://api.uberllm.dev/v1/video/generations \
  -H "Authorization: Bearer $UBERLLM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "fal-ai/veo3.1/fast", "prompt": "a red ball bouncing on green grass",
       "duration_seconds": 8, "resolution": "720p", "audio": true}'

No UberLLM fee on requests. Adding credit by card costs 5.5% + $0.50.

Get an API keyQuickstart

Questions

Can I still call Veo 3.1 with a Gemini API key?

No. Google shut down veo-3.1-generate-preview, veo-3.1-fast-generate-preview and veo-3.1-lite-generate-preview on the Gemini API on October 22, 2026, and points users to Gemini Omni Flash or the GA models on its enterprise platform.

Why are kie.ai and APIMart cheaper?

They price Veo per video rather than per second, the same for 4, 6 or 8 seconds. APIMart’s cheapest prices are on what it calls its “Ext” channel; its official channel is billed per second. We list what each page says.

What does a failed Veo video cost on UberLLM?

Nothing. A video that fails or runs out of time is not billed, and the reply to the request states the most the job can cost.

Are there other fees on UberLLM?

No UberLLM fee on requests. Adding credit by card costs 5.5% + $0.50.