What does the cost Rius shows mean when my team uses a Claude subscription?

Rius shows the cost of each Claude Code session as what its tokens would cost at API list rates, so on a Pro or Max subscription it measures usage, not what you pay. Use it to compare sessions, developers and models, and to see who drives usage. Your subscription bill is a separate figure that Rius does not read or reproduce.

How it works

Rius computes cost itself from the spans. The plugin reports the model and the token counts for each model call, and Rius multiplies them by that model’s rates. You never send a price.

• API rates, not plan rates. Prices come from LiteLLM’s public model-price catalog, in US dollars. A flat monthly subscription has no per-token price, so the figure is an estimate of what the same usage would cost at list prices.
• Standard rates only. Rius does not apply long-context tiers, batch, flex or priority pricing, or any discount you negotiated. Treat every figure as a list-price estimate, not an exact or billed amount.
• Cache tokens count. Cache reads and cache writes are priced at their own rates, so a session that leans on the prompt cache is not overstated at the full input rate.
• Empty is not zero. If a model is not in the catalog, or a call reported no token usage, its cost is left empty. An unpriced call never shows as $0.
• Cost is fixed when the span is written. A later price change does not rewrite past calls.

The full pricing method is in Track LLM and agent cost.

Set it up

1. Each developer installs the plugin in a terminal. The plugin needs Python 3.9 or later.

claude plugin marketplace add glassflow/rius-coding-agents && claude plugin install rius@rius-coding-agents

2. Each developer runs /rius:login, signs in on the browser page, picks the shared team workspace and clicks Connect. Use one workspace for the whole team so cost adds up in one place.
3. In each repo you want traced, run /rius:enable-here. Tracing is off until each developer turns it on for a folder. This mode sends the span tree, models, tokens, cost and timing, and no prompts or replies. Run /rius:status to confirm; the first line reads Rius tracing: on.
4. Start Claude Code from inside the enabled repo, not from a parent folder, so the whole session is traced.
5. Open the Users page and sort by Cost to see which developers use the most. The window is 7 days by default. With plugin 0.6.2 or later, each developer appears under the email they signed in with.

The team guide Track Claude Code usage and cost across a team covers workspace invitations and the other views in full.

What you get

• A usage measure per developer. The Users page lists each developer with their traces, failed traces and cost for the selected window, so you can see who drives usage on the subscription.
• Cost per session and per call. One Claude Code session is one trace. The waterfall shows tokens and cost on each model call, so you can go from a session that cost $12 to the exact step that used the tokens.
• Subagent cost. Each subagent is nested under the call that started it, with its own model calls, tokens and cost.
• A claude-code agent view. The Agents page shows Claude Code as an agent named claude-code, with its traces, errors and cost for the window.
• Questions in plain language. The plugin brings the Rius MCP server, so you can ask Claude which sessions cost the most in the last 7 days and what the tokens went on.

Sessions count on the Users and Agents pages once they have ended. A session that never ends cleanly, because Claude Code crashed or was killed, is not counted there. Sessions sent with an existing API key instead of /rius:login carry no developer and land in the anonymous row.

When something else fits

• Read the Rius figure as relative usage, such as which developer, repo or model used more. For what your team pays for Claude Pro or Max, use your Anthropic billing, which Rius does not see.
• Cursor sessions (beta) carry no token counts or cost, because Cursor’s hooks do not report usage, so no cost figure appears for them. Codex (beta) reports tokens and cost.
• If you pay per token through the API, the Rius figure is a list-price estimate, and your invoice can still differ because of discounts or special pricing tiers.

Related