OpenAI-compatible API

A small, health-gated model catalog.

Use standard OpenAI clients. Each key can be limited by model, spend, expiry, RPM, TPM, and concurrent requests. Routes that stop returning usable output are withheld automatically.

BASE URL

https://api.tensorapi.cloud/v1

Checking API

Live models

USD per million tokens. The API catalog includes only routes that passed the current output and availability gate.

Model	Input	Cached input	Output	Status
Loading catalog…

Quick start

Replace the key and select an ID returned by /v1/models.

curl

curl https://api.tensorapi.cloud/v1/chat/completions \
  -H "Authorization: Bearer $TENSORAPI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kimi-k3",
    "messages": [{"role":"user","content":"Hello"}]
  }'

Python

from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.tensorapi.cloud/v1",
    api_key=os.environ["TENSORAPI_KEY"],
)

response = client.chat.completions.create(
    model="kimi-k3",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Dashboard

Admins invite users. Users set their own password, create and revoke their keys, and inspect only their own usage and spend.

AccessOpen admin.tensorapi.cloud/ui and sign in with your invited account.

KeysCreate project-specific keys instead of reusing one credential everywhere.

LimitsSet model permissions, expiry, budget, RPM, TPM, and concurrency when issuing a key.

LogsUsage metadata is retained for accounting; prompt and completion bodies are disabled.