The OpenAI API you already use.

VXAI is OpenAI-compatible, so your existing SDKs work. Set the base URL and a vxai_ key, and you're routing to Anthropic, OpenAI, or self-hosted models in minutes.

Point any OpenAI SDK at VXAI

# any OpenAI SDK — just change base_url + key
from openai import OpenAI
client = OpenAI(base_url="https://vxai.veeexa.com/v1", api_key="vxai_...")

resp = client.chat.completions.create(
    model="claude-sonnet-4-5",
    messages=[{"role": "user", "content": "Hello from VXAI"}],
    stream=True,
)
for chunk in resp:
    print(chunk.choices[0].delta.content or "", end="")

Or call it with curl

# chat completions (streaming)
curl https://vxai.veeexa.com/v1/chat/completions \
  -H "Authorization: Bearer vxai_..." \
  -d '{"model":"gpt-4o-mini","stream":true,"messages":[{"role":"user","content":"hi"}]}'

# embeddings
curl https://vxai.veeexa.com/v1/embeddings \
  -H "Authorization: Bearer vxai_..." \
  -d '{"model":"text-embedding-3-small","input":"embed me"}'