Up to 80% off sitewide
Compared with official reference prices (Fable models included)
Compare input, output and cache rates with official reference prices. Savings vary by model and group; consult current model pricing for applicable rates.
KeyVia
One API for the world’s leading AI models
Native model quality (full power)
No downgrades. No cutbacks. Verifiable.
Native model quality with no downgrades or cutbacks, and verifiable results.
Reliable global routing
High-concurrency API gateway
Access Anthropic, OpenAI and Gemini models through a unified high-concurrency API gateway for coding tools, applications and teams. See the catalog and setup guides for available models and protocols.
Per-request billing
Usage, latency and cost for every request
Review models, token usage, response latency and costs per request in the console. Filter records by API key and date to reconcile AI API spending across your projects and team.
Models and pricing
Find the right AI model for your workflow, with usage-based pricing.
| Model | Discount | Input | Output | Cache read | Cache write |
|---|---|---|---|---|---|
Anthropic: Claude Fable 5.2 anthropic/claude-fable-5-2 | 80% off | $1.60/M $8.00/M | $8.00/M $40.00/M | $0.04/M $0.20/M | $2.00/M $10.00/M |
Anthropic: Claude Opus 5.5 anthropic/claude-opus-5-5 | 80% off | $0.80/M $4.00/M | $4.00/M $20.00/M | $0.04/M $0.20/M | $1.00/M $5.00/M |
Anthropic: Claude Sonnet 5.5 anthropic/claude-sonnet-5-5 | 80% off | $0.40/M $2.00/M | $2.00/M $10.00/M | $0.02/M $0.10/M | $0.50/M $2.50/M |
OpenAI: GPT 6 Astra openai/gpt-6-astra | 80% off | $2.00/M $10.00/M | $10.00/M $50.00/M | $0.20/M $1.00/M | $2.50/M $12.50/M |
OpenAI: GPT 6.1 Sol openai/gpt-6.1-sol | 80% off | $0.40/M $2.00/M | $2.00/M $10.00/M | $0.02/M $0.10/M | $0.50/M $2.50/M |
OpenAI: GPT 6.1 Terra openai/gpt-6.1-terra | 80% off | $0.20/M $1.00/M | $1.00/M $5.00/M | $0.01/M $0.05/M | $0.25/M $1.25/M |
All models and pricing
Explore the full catalog, pricing tiers and billing details in the model directory.
FAQ
Questions?
Find answers here
See the documentation for setup steps and advanced configuration guides.
Read Setup DocsHow does KeyVia compare with using providers directly?
Access multiple model providers through one API, with USD payments and request-level usage and charges in the console.
How is pricing calculated?
Billing is based on actual token usage. The model catalog lists each model’s unit prices alongside official reference prices, so you can compare rates before paying.
Should I choose pay-as-you-go or a subscription plan?
Choose pay-as-you-go if your usage varies. For frequent daily use, a subscription plan is often more cost-effective. You can switch between the two options at any time.
Does my account balance expire?
No. Pay-as-you-go balance has no expiration date and will never be cleared. Use it whenever you need.
Which clients and tools are supported?
Tools that support the Anthropic or OpenAI protocols and a custom Base URL can connect, including Claude Code, Codex CLI, Gemini CLI, Cursor, Cline, Cherry Studio, and official SDKs.
Can I use one account on multiple computers?
Yes. Create a separate API key for each computer or tool to track usage independently and revoke a compromised key without affecting the others.