View consumption costs directly on the conversation page

Key features

• Single conversation viewThe official status line at the bottom of the conversation page (after "First token average... · ... tok/s") is directly embedded with the consumption cost of this session. Click to open the detailed pop-up window per request (time/model/cache miss/cache hit/output/cost, the latest is at the top).
• Option B (peak and valley pricing, 2026-08-16T16:00Z = starting from 2026-08-17 00:00 Beijing time): peak hours Beijing time 9–12, 14–18 (UTC 1–4, 6–10), half price during leisure hours
• Overall summary (Settings Plug-in Configuration Web UI Plug-in Token Cost Statistics): Time filter (today/yesterday/last 7 days/last 30 days/this month/last month/custom, up to 30 days) + cost/input/output/cache hit rate statistics card, grouped by model, session, date.
• Pricing statusThe current effective plan, next switching time, and peak hours are clear at a glance.
• v4 is applicable to both models, and the old model remains at parity.