AI models & token usage

The Deputy and the Posse call the Anthropic Claude API with YOUR key — you pay Anthropic directly for what you use. There's no SubnetSlinger subscription and no markup. (Prefer no cloud and no cost? Switch the Deputy's provider to Ollama and run a local model instead — see "Ollama — run the AI locally". The rest of this article is about the cloud models.)

The three models (cheapest → most capable):

• Haiku 4.5 — fastest & cheapest. The DEFAULT. Great for quick lookups and most routine questions.
• Sonnet 4.6 — balanced speed/depth. Good for multi-step investigations.
• Opus 4.8 — deepest reasoning, priciest. Save it for genuinely hard problems.

Pick per chat in the Deputy's model dropdown; it opens on Haiku so you only pay for Opus when you choose it.

How you're billed: per TOKEN, split into input (everything sent — your question + tool results + history) and output (the model's reply). Rough list price, US$ per 1,000,000 tokens (check Anthropic for current rates):

Haiku 4.5 ~ $1 in / $5 out

Sonnet 4.6 ~ $3 in / $15 out

Opus 4.8 ~ $5 in / $25 out

A typical short question is a few thousand tokens — fractions of a cent on Haiku.

The token readout: above the Deputy's composer (and by the Posse's status) a live line shows 'tokens · in N · out N · ~$X this chat' so you can watch spend as you go. It resets on New chat. The Posse blends specialist (Sonnet) + coordinator (Opus) rates for an accurate per-run estimate — a 5-agent sweep costs more than one Deputy question, so run it when you want the thorough pass.

Keeping cost down: SubnetSlinger auto-trims old tool-result payloads from a long conversation (~30k-token cap) so the input doesn't balloon turn after turn; start a New chat to drop context entirely; and stay on Haiku until a problem actually needs Sonnet/Opus. Tool runs themselves (ping, SNMP, etc.) are local and free — you only pay for the model's reasoning over them.