Every model, lowest rate.
Fast edge routing with honest per-1M token prices passed straight from each provider. Zero markup on your first $5 — and instant failover built in.
Showing 25 of 25 · prices refresh every 5 minutes
Live in 30 seconds.
from openai import OpenAI client = OpenAI( base_url="https://api.vigas.ai/v1", api_key="sk-vg-v1-…", ) resp = client.chat.completions.create( model="anthropic/claude-opus-5.5", # or "vigas/auto" messages=[{"role": "user", "content": "Hello"}], )
Sign in to see your balance and keys.
Automatic fallbacks
If a provider is down or rate-limited, the request moves to the next one — same response format.
Fastest edge routing
Route requests by latency, lowest price or Arena quality, or let vigas/auto pick the fastest healthy endpoint.
Lowest pass-through prices
Prepaid credits across every provider with $0 markup on your first $5 and zero hidden surcharges.
Usage you can read
Per-key spend, tokens and daily usage — in your dashboard with optional monthly caps.
Questions developers actually ask.
Is Vigas compatible with the OpenAI SDK?
Yes. Point base_url at https://api.vigas.ai/v1, use your Vigas key, and pick any model ID from this page.
How is pricing calculated?
You pay each provider’s per-token price, shown above. The first $5 of credits has no markup; after that a small platform fee is shown before you pay.
What happens when a provider goes down?
With fallbacks on, Vigas retries the same request on the next best provider and tells you which one answered.
Which tools and SDKs work with a Vigas API key?
Any client supporting OpenAI-compatible endpoints works out of the box — including the official Python and TypeScript SDKs, Cursor, Cline, OpenCode, and production backends.