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

One API for text, image and video models (October 2026)

Updated October 31, 2026, 16:00 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, one UberLLM key reached 248 chat, 79 image and 104 video models and 15 music models that could take requests, through OpenAI-style endpoints: chat completions, image generations and video generations. OpenRouter, CometAPI, 01Router, EachLabs and AIMLAPI also serve all three with one key; kie.ai, fal and Replicate lead on media and call each model through its own API. They differ on fees, whether failed jobs are charged and whether credit expires.

ProviderText, image, videoOpenAI-style chat endpointFee on topFailed generationsCredit expires
UberLLMYes: 248 chat, 79 image, 104 video, 15 musicYes, plus Anthropic /v1/messagesNone on requests; 5.5% + $0.50 card feeNot chargedNo, while the account is open
OpenRouterYes: 62 image and 31 video models in its listsYes5.5% when buying credit“An attempt that errors is not charged for its model tokens”May expire after one year
kie.aiYes, plus musicPer-model paths, not one endpoint“No hidden fees”“Failed tasks are never charged”Never
CometAPIYes, plus speech and musicYesSays no hidden fees; prices at 80% of officialNot statedNever
01RouterYes, plus Suno musicYesNot statedRefunded automaticallyAfter 365 days
EachLabsYes, plus 3D and voiceYes (each::router)LLMs: list price plus 5.5%Not statedNot stated
AIMLAPIYes, plus music and speechYesNot statedMay be charged when the model provider failsNo time limit on the plan
falImage and video lead; chat only through one router modelThrough one model page onlyNot statedServer errors not charged; refusals may beAfter 365 days
ReplicateYes, by modelNot statedNot stated“If a run fails, we don’t charge you”Prepaid credit is valid for 1 year

UberLLM first, then the services that serve text, image and video through one OpenAI-style chat endpoint, then media-first APIs. An empty cell means we could not confirm it on the provider's own pages.

Each provider, good and bad

UberLLM

Pick it if: Chat models below list price and media models on one prepaid key, with the most a video can cost stated before it runs.

Watch for: Fewer media models than fal or Replicate, and no free tier.

Source: UberLLM terms, media docs (live, October 31, 2026)

OpenRouter

Pick it if: The largest chat catalog, now with image and video endpoints.

Watch for: Chat at list price; credit can expire after a year.

Source: OpenRouter pricing, video models, FAQ (checked 2026-10-30)

kie.ai

Pick it if: Low per-video prices on Veo, Kling and Seedance.

Watch for: Each model has its own API path.

Source: kie.ai pricing, kie.ai docs (checked 2026-10-30)

CometAPI

Pick it if: Prices at 80% of official, across 500+ models by its own count.

Watch for: Minimum top-up $10; failed-job billing not stated.

Source: CometAPI pricing (checked 2026-10-30)

01Router

Pick it if: Credit packs with automatic refunds for failed tasks.

Watch for: Smallest pack $10; credit expires after a year.

Source: 01Router (checked 2026-10-30)

EachLabs

Pick it if: Media workflows plus an LLM router, no minimum.

Watch for: LLMs cost list price plus 5.5%.

Source: EachLabs pricing (checked 2026-10-30)

AIMLAPI

Pick it if: A long model list with no subscription.

Watch for: $20 minimum top-up; a provider-side failure may still be charged.

Source: AIMLAPI plans (checked 2026-10-30)

fal

Pick it if: The most media models and day-one launches.

Watch for: Not a chat API; credit expires after a year.

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

Replicate

Pick it if: Thousands of models with versioned endpoints.

Watch for: Per-model APIs rather than one chat endpoint.

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

How to switch to UberLLM

The same key makes chat, images and video on UberLLM:

python
from openai import OpenAI

client = OpenAI(base_url="https://api.uberllm.dev/v1", api_key="ull_your_key_here")

chat = client.chat.completions.create(
    model="deepseek/deepseek-v4-flash",
    messages=[{"role": "user", "content": "Write a 10-word video prompt."}],
)
img = client.images.generate(model="venice/nano-banana-pro", prompt="a red apple on a white table")

# Video is a job: POST /v1/video/generations, then poll the job (see /docs/media).

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

Questions

Which API serves GPT, Claude, Veo, Kling and image models with one key?

UberLLM, OpenRouter, CometAPI, 01Router, EachLabs and AIMLAPI each serve chat, image and video models with one key through OpenAI-style endpoints. kie.ai, fal and Replicate cover many media models but use per-model APIs for chat.

Are failed videos charged?

On UberLLM, kie.ai, 01Router and Replicate a failed generation is not charged, by their own pages. fal does not charge server errors but may charge refusals; AIMLAPI may charge when the model provider fails.

Does UberLLM credit expire?

No. Credit does not expire while your account is open, and unused credit can be refunded on request within 30 days of purchase.

Are there other fees on UberLLM?

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