Skip to main content

Other tools and SDKs

Any tool that lets you set a custom OpenAI-compatible base URL works with Standard Compute. Use these values:

SettingValue
Base URLhttps://api.stdcmpt.com/v1
API keyYour key from the dashboard
Modelstandardcompute

Cline​

Standard Compute is a built-in provider in Cline 4.1.25 for VS Code and JetBrains, Cline CLI 3.0.72 and Cline Desktop 0.0.47 (all released October 16, 2026).

In the extension, open Cline's settings and pick Standard Compute under API Provider. Enter the base URL, paste your key under OpenAI Compatible API Key, and keep standardcompute as the Model ID. In the CLI, run:

cline auth -p standardcompute -k "$STANDARDCOMPUTE_API_KEY" -m standardcompute

On Cline 4.1.24 or CLI 3.0.71 and older, set API Provider to OpenAI Compatible instead and enter the same base URL, key and Model ID. See Cline's docs.

Kilo Code and ZooCode​

Open the extension's settings and choose the OpenAI Compatible provider. Enter the base URL and your API key, and set the model to standardcompute. See Kilo Code's docs.

Roo Code shut down on May 15, 2026 and its repository is archived (sunset notice). The same three settings work in Cline, in Kilo Code and in ZooCode, the community fork that keeps Roo's modes, and in an existing Roo Code install while you move.

Aider​

export OPENAI_API_BASE=https://api.stdcmpt.com/v1
export OPENAI_API_KEY=YOUR_API_KEY
aider --model openai/standardcompute

You can also set these in ~/.aider.conf.yml. See Aider's docs.

Continue​

Continue is no longer maintained: its team joined Cursor and the repository is read-only. An existing install still works; for a new setup, use Cline or Kilo Code. In your Continue config, add a model with provider openai, set apiBase to the base URL, apiKey to your key and model to standardcompute. See Continue's docs.

Zed​

In Zed's settings.json, add an OpenAI-compatible provider under language_models with api_url set to the base URL. Add your API key and use standardcompute as the model. See Zed's docs.

SDKs and HTTP clients​

The official OpenAI SDKs work unchanged; set the base URL and key. See the Quickstart for Python, Node.js and curl examples. For the Vercel AI SDK, see Vercel AI SDK.

Libraries built on the OpenAI client, such as LangChain, LiteLLM and Instructor, work the same way. For example, with LangChain:

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="standardcompute",
    base_url="https://api.stdcmpt.com/v1",
    api_key="YOUR_API_KEY",
)

print(llm.invoke("Summarize this in one sentence: ...").content)

For long agentic requests, set client timeouts to at least 60 seconds.