Which API should you use with nanobot?

For nanobot, the self-hosted personal agent from HKU's Data Intelligence Lab, the best API for an always-on setup is a cost-optimized OpenAI-compatible provider: add a standardcompute provider with apiBase https://api.stdcmpt.com/v1 to ~/.nanobot/config.json (or Settings → Models → New preset → Add provider) and a preset with the model standardcompute, and Standard Compute's router sends each step to the cheapest model that can do it (Claude Fable 5.1 or GPT-6 Astra only when it counts), on a plan from $19 a month with a $20 compute budget. OpenRouter and direct provider keys bill per token, and a local model through Ollama costs nothing per token. nanobot itself is free and MIT-licensed.

Light or occasional use

For a nanobot that answers a few messages a day, a pay-as-you-go key on an efficient model will likely cost the least. A local model through Ollama works if your machine stays on.

Daily use, cost-optimized

Once nanobot runs scheduled tasks, chat channels and tool calls around the clock, a Standard Compute provider sends each step to the cheapest model that can do it, and for every model we vet providers for price, uptime and speed, with automatic fallback. Opt-in Europe or US Compliance rules apply per key, team or company, and we don't train on your prompts or store them. The bill stays at the monthly plan price: requests stop with HTTP 402 when the compute budget is used, so a runaway loop cannot bill you on top.

API and subscription options for nanobot

Prices and allowances can change. nanobot's provider and preset format and its ${VAR} secret references come from its providers and configuration docs, read on October 15, 2026. nanobot: providers and models · nanobot: configuration · nanobot repo. Spot an error? Tell us.

Add Standard Compute to nanobot

{
  "providers": {
    "standardcompute": {
      "apiKey": "${STANDARDCOMPUTE_API_KEY}",
      "apiBase": "https://api.stdcmpt.com/v1"
    }
  },
  "modelPresets": {
    "primary": {
      "provider": "standardcompute",
      "model": "standardcompute",
      "maxTokens": 8192,
      "contextWindowTokens": 200000
    }
  },
  "agents": { "defaults": { "modelPreset": "primary" } }
}

Merge this into the config nanobot onboard created, export STANDARDCOMPUTE_API_KEY in the environment that starts nanobot, and restart it. In the WebUI the same lives under Settings → Models → New preset → Add provider.

nanobot questions

Why does nanobot's custom provider fail without a base URL?

nanobot's custom provider does not infer a default base URL, so you must set apiBase, including the version path: https://api.stdcmpt.com/v1 for Standard Compute. Give each OpenAI-compatible endpoint its own key under providers (for example standardcompute) and use that same key as the preset's provider.

Can nanobot use an Anthropic-compatible endpoint?

Yes. For a named custom provider that serves Anthropic Messages, set providers.<name>.api.supportedApis to ["anthropic_messages"]; for an Anthropic-compatible proxy, nanobot's docs say to keep the provider as anthropic and override apiBase. The OpenAI-compatible setup above is the simpler one, and Standard Compute's routing works the same way on both.

How do I keep my API key out of nanobot's config.json?

Use a ${VAR} reference, such as "apiKey": "${STANDARDCOMPUTE_API_KEY}". nanobot resolves it from the environment at startup, so start nanobot from the same environment that sets the variable.