One API for frontier AI models.
State-of-the-art models from the leading AI labs.
curl https://hxa.net/v1/chat/completions \ -H "Authorization: Bearer $HXA_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "MODEL_ID", "messages": [{"role": "user", "content": "Hello"}] }'
Models Indicative
A few of the models we serve. USD per 1M tokens, input / output. Volume pricing — contact us.
| Model | List price | hxa.net |
|---|---|---|
| GPT-6.1 SolOpenAI | $2 / $10 | $1.40 / $7 |
| Claude Opus 5.5Anthropic | $4 / $20 | $2.80 / $14 |
| GPT-6 AstraOpenAI | $10 / $50 | $7 / $35 |
How it works
Already using an AI API? You only change the base URL and the key.
Create an account
Sign up with your email, then create an API key in the console.
Add credit
Contact us to top up your balance. Online payment is coming soon.
Change the base URL
Point your existing SDK at hxa.net and keep the rest of your code.
Quality and reliability
Built for teams that run AI models in production.
Uptime
99.9%
Uptime SLA on the API.
Latency
< 50 ms
Added gateway latency (p50) on top of the model's own response time.
Frontier coverage
Latest SOTA
The newest state-of-the-art models, kept up to date.
FAQ
Do I need to change my code?
No. Set the base URL to hxa.net and use your hxa.net key.
How am I billed?
Per token, in USD. Usage is deducted from your account credit per request. Contact us for volume pricing.
Which models are available?
See the pricing page for the full list and the price of each model.
Is there a free trial?
Contact us for trial credit to evaluate the models before you add funds.
How do I get support?
Every account has a named contact. Talk to us about capacity and rate limits.
Are the models genuine?
Yes — official models from the original providers, never substituted or diluted. With the optional on-premise router, you can verify every request yourself.
How is my data privacy handled?
With the optional on-premise router, your requests go straight to the provider — never through our servers.
Talk to our team
Tell us your models and volume. We will get back to you with pricing.