Per-token, matched to source. Not a credit pack.
WorldClaw sells AI token credits in four fixed pack tiers (US$9.9 / 99 / 999 / 9999), settled in USD1 or by locking WLFI tokens. OminiGate bills the real tokens your request consumes — at rates aligned 1:1 with the upstream provider, on pay-as-you-go top-ups, across text, image, and video.
- PricingPer-token, 0% markup
- ModalitiesText · Image · Video
- Top-upPay-as-you-go, no packs
Black Forest Labs
Google
Moonshot
Pixverse
Recraft
Qwen
Z.AIFour reasons to choose OminiGate
Real per-token billing, not token credit packs
WorldClaw sells "AI token credits" in four fixed packs, with illustrative rates roughly 30% below upstream — explicitly not guaranteed pricing. OminiGate bills the exact tokens your request consumes at rates aligned 1:1 with the upstream provider, with no credit-to-token conversion layer.
Image and video generation, not just chat
OminiGate covers mainstream image (GPT Image, Seedream, Flux) and video (Veo, Seedance, Kling, Pixverse) generation alongside the text LLM lineup — under the same key, the same balance, the same dashboard.
Mainstream cross-provider lineup
OminiGate ships full SKUs from OpenAI, Anthropic, Google, DeepSeek, xAI, Moonshot, MiniMax, ByteDance, Black Forest Labs and more. Switch providers by changing the model slug, with the same SDK and the same key.
Pay-as-you-go from your balance
OminiGate runs on a top-up balance — your balance funds every call across every modality, billed per token at the source rate. No fixed pack tiers, no credit conversion.
Beyond chat. Image and video, same key.
OminiGate covers text, image, and video today — funded by one balance, tracked in one dashboard.
- Mainstream cross-provider text LLMs (OpenAI, Anthropic, Google, DeepSeek, xAI, Moonshot and more)
- Image generation (GPT Image, Seedream, Flux) and video generation (Veo, Seedance, Kling, Pixverse)
- One pre-paid balance funds every modality, with per-modality cost in real time
- Voice and audio on the roadmap — your existing key will work on day one
sk-omg-•••••••Built for production workloads
Direct per-token billing
Pay for the tokens your request consumes — not credits that convert to calls. Every model's per-token and per-call rate is published on the pricing page, matched to the upstream provider.
Image and video endpoints
Dedicated routes for image generation and video — same auth, same balance, same dashboard.
Drop-in OpenAI / Anthropic SDK
Keep the client library you already use. Change one base URL — chat completions, messages, tool calls all keep working.
Real-time usage events
Per-request cost and latency visible right after the response — no delay, no sampling.
Move from WorldClaw in three steps
Need a hand? The team replies to integration emails directly. Contact us (contact@ominigate.ai)→
Get your OminiGate key
Sign up with email, verify, copy the sk-omg- key from the dashboard. Under a minute.
# Dashboard → API Keys → New
sk-omg-xxxxxxxxxxxxxxxxTop up your balance
Pay-as-you-go from your balance — no fixed pack tiers. Your balance funds every call across every modality.
- WorldClaw tokens: $9.9 / $99 / $999 / $9999 packs
+ OminiGate: pay-as-you-go from your balance
# No fixed pack tiers, no credit conversionCall any model at upstream rates
Point the OpenAI or Anthropic SDK at OminiGate. Per-token rates match the source — verify on the pricing page before you ship.
# pricing page lists per-model rates
client.chat.completions.create(
model='openai/gpt-5.4-pro', # source-matched rate
messages=[...],
)Open SDKs. Open prices.
Bring the OpenAI or Anthropic client library you already ship. Point it at OminiGate and a mainstream cross-provider catalog lights up — at rates that match the source provider 1:1.
OpenAI-compatible chat completions
Chat completions, tool calls, streaming, vision — all follow the OpenAI shape.
Anthropic-compatible messages
/v1/messages with system, tools, streaming, extended thinking — Claude SDK plugs in unchanged.
Image and video endpoints
Dedicated routes for image and video generation — same key, same balance.
Real-time usage events
Per-request cost and latency visible right after the response — no delay, no sampling.
from openai import OpenAI
client = OpenAI(
base_url="https://api.ominigate.ai/v1",
api_key="sk-omg-...",
)
# Per-token rates listed openly — verify before shipping
resp = client.chat.completions.create(
model="anthropic/claude-opus-4.6",
messages=[{"role": "user", "content": "Hi"}],
)curl https://api.ominigate.ai/v1/images/gpt/text2img \
-H "Authorization: Bearer sk-omg-..." \
-d '{
"model": "openai/gpt-5-image",
"messages": [{"role": "user", "content": "cat reading a book"}],
"image_config": {"aspect_ratio": "3:4"}
}'Source rates. Zero markup.
Every model on OminiGate is listed at its upstream provider price — what you pay is what the source charges, billed against the exact tokens your request spends. No credit conversion layer between your usage and the bill. Volume tiers structured openly.
See full pricing→Real models. Real prices.
What you gain when the bill ties to the tokens you actually spend and the catalog covers text, image, and video generation under one key.
Matched to the source, billed on actual usage
Every model's per-token and per-call price is published on the pricing page, aligned 1:1 with the upstream provider. The bill ties to the actual tokens spent — no credit-to-token conversion layer between your usage and what the source charges.
- Mainstream cross-provider text LLM lineup under one key
- Image and video generation endpoints today
- Pay-as-you-go balance — no fixed pack tiers
- Voice and audio on the product roadmap
Gateway markup
Published rates match the upstream. Volume tiers negotiated openly.
messages + chat completions
Point either Anthropic or OpenAI SDK at OminiGate.
Questions developers ask before switching
Real models. Real prices.
Sign up, top up your balance, and call text, image, or video models through the same OpenAI- or Anthropic-compatible endpoint.