An unlimited API for agents: pay for streams, not steps

Why agents get expensive

Every agent step is a new model call that resends instructions, step history and file contents. On a long task the same context is paid for dozens of times.

A Tenzo Cloud subscription removes that link: the price is fixed per stream and the number of tokens that pass through the agent does not matter. Only concurrent requests are limited.

What you can run

• Claude Code, Codex CLI, OpenCode and Cline in long sessions
• Your own agents on LangChain, LangGraph and the OpenAI SDK
• Overnight runs: test generation, refactoring, migrations
• Support bots and n8n automation

How many streams an agent needs

A stream is one concurrent request. A sequential agent is fine with one or two. If it runs parallel branches or helpers, take 4, our recommendation for agent work. Streams for one model stack and you can add more any time.

Connect an agent in Python

from openai import OpenAI

client = OpenAI(base_url="https://tenzo-cloud.com/v1", api_key="tc-your-key")

stream = client.chat.completions.create(
    model="glm-5.3",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")