Back

Standard Compute API Documentation

Connect Standard Compute to any service supporting OpenAI-style API endpoints.

Base URL: https://api.stdcmpt.comAuth: Authorization: Bearer <API_KEY>Endpoints: /v1/completions · /v1/responsesUnlimited (fair use)Model: standardcompute
API v3.1 · Last updated June 2026
In a nutshell

Standard Compute is built to be a plug-and-play AI API for automations. If your automation platform can send an HTTP request, it can use Standard Compute — no special SDKs required.

Best for
Agent automations, tool-calling workflows, summarization, extraction, classification, and any “AI as a step” pipeline.
What you get
One API key with unlimited compute. Our routing algorithm handles model selection across top-tier providers — you just send requests.
Compatibility
OpenAI-compatible endpoints. Swap your base URL and API key — existing workflows keep working.

How model routing works

Every request hits the Standard Compute gateway — a single, fully OpenAI-compatible endpoint. From there, Standard Compute's intelligent routing algorithm requests the best-fit model across top-tier providers — reaching for the most advanced models when the work calls for it. You integrate once against one Standard Compute endpoint and one model id, and we do the rest.

Standard Compute lifecycle
Your app
any OpenAI client
standardcompute
the only model
Router
scores the task
Best model
matched to task
Response
OpenAI shape
The Router matches each request to the right tier:
Deep reasoningHardest multi-step problemsDeep analysisLong context
Coding & agenticComplex codeTool-heavy agent loops
BalancedEveryday tasksStrong quality + speed
Fast & lightSimple + high-volumeLatency-sensitive
One gateway
Point any OpenAI SDK or tool at https://api.stdcmpt.com/v1. No custom SDKs — override the base URL and key, and existing code works unchanged.
Smart routing
Always request the model standardcompute. Our routing algorithm reads each request and matches it to the ideal model — reaching for the most advanced models when a task needs deeper reasoning, and fast ones for simple steps. You always get the best fit, no model pinning required.
Normalized I/O
Requests and responses follow OpenAI schemas across /chat/completions, /completions, and /responses, with up to 200K context — your existing parsers keep working.

Authentication

All requests use a customer API key in the Authorization header:

Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

Tip: store the key in your automation platform's “Secrets/Credentials” store (recommended) rather than directly in node/module configuration.

Core Endpoints

EndpointUse it forTypical output
POST /v1/completionsFast prompt to completion workflows. Great for “one-shot” tasks.Completion text under choices[0].text
POST /v1/responsesRicher responses and structured output workflows.Convenient text field such as output_text

Example: /v1/completions

POST https://api.stdcmpt.com/v1/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "standardcompute",
  "prompt": "Summarize this message in 1 sentence:\n{{text}}",
  "max_tokens": 120,
  "temperature": 0.3
}

Example: /v1/responses

POST https://api.stdcmpt.com/v1/responses
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "standardcompute",
  "input": "Extract key action items from this:\n{{meeting_notes}}"
}

General Best Practices

  • Keep temperature low (0.2–0.5) for predictable automation steps.
  • Prefer structured prompts (e.g. “Return JSON with fields: …”) when you need consistent mapping.
  • Log request IDs (if included in responses) to simplify support and debugging.
  • Use /responses when you want richer or more structured output; use /completions for simpler text generation.

Troubleshooting

401 / Unauthorized

  • Verify the header is exactly: Authorization: Bearer YOUR_API_KEY
  • Confirm you are not sending the key as a query parameter.

400 / Bad Request

  • Ensure JSON is valid and Content-Type: application/json is set.
  • Confirm you are using the correct field names for the endpoint (prompt vs input).

n8n OpenAI nodes don't allow Base URL

  • Use the n8n HTTP Request node method (always supported).

Hermes Agent Integration Recommended agent

Hermes is an autonomous AI agent that ships with a large set of built-in skills and tools (web search, browser automation, file operations, text-to-speech, cron jobs, and more). It connects to any OpenAI-compatible endpoint, so Standard Compute drops straight in as the model provider.

Recommended: The Hermes Agent Setup page in your dashboard has a copy-paste settings card and a full 10-minute setup guide. The steps below are a condensed reference.

Before you start

Hermes runs on Linux or macOS (use WSL2 on Windows). You'll need git, curl, python, node.js, and your Standard Compute API key. A small VPS or an older machine is enough when the model runs remotely.

Step 1 — Install Hermes

Run the install script, reload your shell, and verify the install:

curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash
source ~/.bashrc
hermes --version

Step 2 — Run the setup wizard

Start hermes setup and choose Full setup — you already have your Standard Compute key, so you don't need the Quick Setup trial inference. When prompted for a provider, select Custom OpenAI-compatible endpoint and paste your Standard Compute connection:

Base URL:  https://api.stdcmpt.com/v1
API Key:   YOUR_STANDARDCOMPUTE_API_KEY
Model:     standardcompute

Hermes auto-detects the model's capabilities and recommends a setup type (responses, chat-completions, etc.) — accept the recommendation unless you know otherwise.

Step 3 — Backend, tools & messaging

  1. Choose the local terminal backend to run Hermes directly in your environment (Docker is also an option).
  2. Enable only the tools you need. Giving a weaker model every tool makes it more likely to choose the wrong one or loop.
  3. Optionally connect Telegram: message @botfather, send /newbot, and paste the token into the wizard.

Step 4 — Start and test

Launch the chat with hermes, then confirm the wiring and a simple tool task:

Hello. Which provider and model are you using?

Write a bash command that shows disk usage.
Looping or wrong-tool issues? On Standard Compute the model isn't the bottleneck — our router serves a capable model for each request and won't cap you mid-loop — so tighten the setup instead: reduce enabled tools, write clearer instructions, confirm the responses setup type, or lower max_turns (e.g. 30). If the hermes command isn't found, run source ~/.bashrc (or ~/.zshrc). hermes doctor diagnoses most problems.

Hermes use cases

  • A always-on personal agent you message from Telegram, backed by unlimited Standard Compute inference
  • Scheduled cron jobs that research, summarize, and report without per-token cost worries
  • Skill-driven automations (calendar, web scraping, file ops) running on a small VPS or spare machine

OpenClaw Integration

OpenClaw is an open-source AI agent framework that lets you run autonomous agents with pluggable model providers. By registering Standard Compute as a provider in OpenClaw’s config, every agent session automatically routes through the Standard Compute API — no per-agent setup needed.

Recommended: The fastest way to configure this is through the OpenClaw Setup page in your dashboard. It handles validation and writes the config for you. The manual steps below are provided for reference or environments where the setup page isn’t available.

Why configure a provider?

OpenClaw resolves model requests through a provider chain. By adding Standard Compute as a named provider and setting it as the default, all agent calls — primary model, sub-agents, image handling, and heartbeat pings — are routed to api.stdcmpt.com without touching individual agent definitions. If Standard Compute is ever unreachable, OpenClaw can fall back to any previous provider you had configured.

Before you start

Back up your existing config so you can roll back if anything goes wrong:

cp ~/.openclaw/openclaw.json ~/.openclaw/openclaw.json.bak

Step-by-step (manual config)

Open ~/.openclaw/openclaw.json in your editor. Use OpenClaw’s config command or API where possible rather than free-hand editing — it validates the schema for you.

  1. Set the provider merge mode. This tells OpenClaw to merge your new provider entry with the existing list instead of replacing it, so nothing else breaks.
    "models": {
      "mode": "merge",
      ...
    }
  2. Register the Standard Compute provider. Add (or replace) the standardcompute key under models.providers. This tells OpenClaw where to send requests and what capabilities the model exposes.
    "models": {
      "providers": {
        "standardcompute": {
          "baseUrl": "https://api.stdcmpt.com/v1",
          "apiKey": "<YOUR_API_KEY>",
          "api": "openai-responses",
          "models": [
            {
              "id": "standardcompute",
              "name": "Standard Compute",
              "reasoning": false,
              "input": ["text", "image"],
              "cost": { "input": 0, "output": 0, "cacheRead": 0, "cacheWrite": 0 },
              "contextWindow": 200000,
              "maxTokens": 8192
            }
          ]
        }
      }
    }

    The api field is set to openai-responses because Standard Compute exposes an OpenAI-compatible responses endpoint. Costs are zero because Standard Compute billing is handled at the account level, not per-token.

  3. Point the defaults to Standard Compute. These four keys control which model OpenClaw picks for each role. Setting them all to standardcompute/standardcompute (provider/model) means every agent component uses the same endpoint.
    "agents": {
      "defaults": {
        "model":      { "primary": "standardcompute/standardcompute" },
        "imageModel": { "primary": "standardcompute/standardcompute" },
        "subagents":  { "model":   "standardcompute/standardcompute" },
        "heartbeat":  { "model":   "standardcompute/standardcompute" }
      }
    }
  4. Preserve fallbacks. If your previous primary model was different (e.g. openai/gpt-4o), move it into the fallback arrays so OpenClaw can still reach it when needed:
    "model":      { "primary": "standardcompute/standardcompute", "fallbacks": ["openai/gpt-4o"] },
    "imageModel": { "primary": "standardcompute/standardcompute", "fallbacks": ["openai/gpt-4o"] }
  5. Disable built-in image media. Standard Compute handles image input through the model itself, so the separate media pipeline should be turned off to avoid duplicate processing:
    "tools": {
      "media": {
        "image": { "enabled": false }
      }
    }
  6. Leave everything else alone. The merge mode (step 1) ensures unrelated config — other providers, tool settings, agent definitions — stays untouched.
  7. Validate before restarting. Run OpenClaw’s built-in validation to catch schema errors before they cause a failed startup:
    openclaw config validate

    Only restart the agent runtime once validation passes. If it reports errors, fix them first — a bad config can prevent OpenClaw from starting entirely.

OpenClaw use cases

  • Autonomous coding agents backed by unlimited Standard Compute inference
  • Multi-agent pipelines where sub-agents and the primary model share one provider
  • Local development setups that need a zero-cost model endpoint for rapid iteration

OpenCode Integration

OpenCode is an open-source terminal coding agent. Because it speaks the OpenAI protocol, Standard Compute drops in either as a provider entry in opencode.json or through plain environment variables.

Recommended: The OpenCode Setup page has a copy-paste settings card with your live API key.

Option A — Provider config (opencode.json)

Add Standard Compute under provider in ~/.config/opencode/opencode.json, then select the model.

{
  "provider": {
    "standardcompute": {
      "npm": "@ai-sdk/openai-compatible",
      "options": {
        "baseURL": "https://api.stdcmpt.com/v1",
        "apiKey": "YOUR_API_KEY"
      },
      "models": {
        "standardcompute": { "name": "Standard Compute" }
      }
    }
  }
}

Option B — OpenAI-compatible mode (environment variables)

OpenCode also reads standard OpenAI environment variables. Add these to your ~/.zshrc (or ~/.bashrc) and run source ~/.zshrc to apply:

export OPENAI_API_KEY=YOUR_API_KEY
export OPENAI_BASE_URL=https://api.stdcmpt.com/v1

Despite the “OpenAI” naming, this is a protocol — not a provider restriction. The endpoint at https://api.stdcmpt.com/v1 serves the model Standard Compute's router selects for each request, behind the stable standardcompute model name.

OpenCode use cases

  • Terminal-native coding sessions backed by unlimited Standard Compute inference
  • Dropping into existing OpenAI-configured projects by swapping two environment variables
  • Scripted/CI coding tasks that need a single zero-config model endpoint

Kilo Code Integration

Kilo Code is an AI coding agent for VS Code. It connects to Standard Compute through its built-in OpenAI Compatible provider — no config files required.

Recommended: The Kilo Code Setup page has a copy-paste settings card with your live API key.

Step-by-step (VS Code)

  1. Open the Kilo Code settings panel in VS Code.
  2. Set API Provider to OpenAI Compatible.
  3. Enter these three values:
Base URL:  https://api.stdcmpt.com/v1
API Key:   YOUR_API_KEY
Model:     standardcompute

Save, and Kilo Code routes every request through Standard Compute.

Kilo Code use cases

  • In-editor refactors, reviews, and feature work without per-token cost worries
  • Teams standardizing on one unlimited endpoint across every developer's IDE
  • Switching from a metered OpenAI key by changing only the Base URL and key

n8n Integration Best experience

n8n is the most flexible platform for custom APIs, and Standard Compute integrates cleanly in two ways:

  • True plug-and-play with n8n's OpenAI tooling (when your n8n version supports a custom Base URL in OpenAI credentials).
  • Universal method using the n8n HTTP Request node (works in all versions).
True plug-and-play (OpenAI nodes): If your n8n OpenAI credentials screen includes a Base URL field, you can simply swap out OpenAI credentials and keep using the built-in OpenAI nodes — no workflow refactor.

Option A — Use n8n OpenAI credentials (plug-and-play)

  1. Go to Credentials in n8n.
  2. Create or edit an OpenAI credential.
  3. Set: API Key: YOUR_STANDARDCOMPUTE_API_KEY and Base URL: https://api.stdcmpt.com/v1
  4. Use the OpenAI/Chat/LLM nodes as usual.
Important: Some n8n versions/nodes do not expose a configurable OpenAI Base URL. If you don't see a Base URL field (or your node ignores it), use Option B below (HTTP Request), which works everywhere.

Option B — HTTP Request node (works in all n8n versions)

Use HTTP Request and call Standard Compute directly:

Method: POST
URL: https://api.stdcmpt.com/v1/completions
Headers:
  Authorization: Bearer {{$env.STDCP_API_KEY}}
  Content-Type: application/json
Body (JSON):
{
  "model": "standardcompute",
  "prompt": "Rewrite this politely:\n{{$json.text}}",
  "max_tokens": 180,
  "temperature": 0.4
}

n8n use cases

  • Summarize incoming support tickets and route them by topic.
  • Classify leads from forms and auto-assign to teams.
  • Generate structured “next steps” from meeting notes.

Zapier Integration

Zapier is fantastic for fast no-code automations. For custom AI APIs, the most reliable approach is to use Webhooks by Zapier (Custom Request).

Action: Call Standard Compute using “Webhooks by Zapier”

Use the Custom Request action to send JSON to Standard Compute.

Step-by-step (Zapier UI)

  1. Create a new Zap in Zapier.
  2. Choose any Trigger app (e.g. Gmail, Google Forms, Slack, Webhook Catch Hook).
  3. Click + to add an Action.
  4. Search for and select Webhooks by Zapier.
  5. In Event, choose Custom Request.
  6. Configure: Method: POST, URL: https://api.stdcmpt.com/v1/responses (or /v1/completions).
  7. Click ContinueTest step to verify you get a JSON response.

Example payload: /v1/responses

POST https://api.stdcmpt.com/v1/responses

{
  "model": "standardcompute",
  "input": "Summarize this in 3 bullet points:\n{{Text from previous step}}"
}

Example payload: /v1/completions

POST https://api.stdcmpt.com/v1/completions

{
  "model": "standardcompute",
  "prompt": "Rewrite this politely:\n{{Text from previous step}}",
  "max_tokens": 200,
  "temperature": 0.3
}

Zapier will expose fields from the JSON response so you can map them into later steps. For example, a completion may be accessible via choices__0__text.

Zapier use cases

  • Google Forms → Standard Compute summary → Slack message
  • Gmail → classify intent → create CRM task
  • Spreadsheet row → generate description → publish draft

Make.com Integration

Make.com integrates cleanly through its HTTP module (“Make a request”) and custom webhooks. You can call Standard Compute in one module and map the response to subsequent steps.

HTTP Module: “Make a request” (recommended)

Use the HTTP app for both /v1/completions and /v1/responses.

Step-by-step (Make.com UI)

  1. Create a new Scenario.
  2. Click the + button to add your first module.
  3. Search for HTTP and select it.
  4. Choose Make a request.
  5. Configure: Method: POST, URL: https://api.stdcmpt.com/v1/completions (or /v1/responses), and add the authorization header.
  6. Click OK, then Run once to test and inspect the output bundle.

Example JSON body: /v1/completions

{
  "model": "standardcompute",
  "prompt": "Turn this into a friendly customer reply:\n{{message}}",
  "max_tokens": 220,
  "temperature": 0.4
}

Example JSON body: /v1/responses

{
  "model": "standardcompute",
  "input": "Extract action items:\n{{meeting_notes}}"
}

Make.com use cases

  • Shopify order → generate confirmation message → email customer
  • Webhook event → extract entities → update Airtable
  • Content idea → generate post draft → schedule in social tool

Implementation Notes & Sources

We keep this guide aligned with platform documentation and community-known behavior. For reference:

Standard Compute · Base API: https://api.stdcmpt.com