Measured harness ledgerPublic result
Qwen3.8 MaxRed Sands v1 — Qwen3.8 Max xhigh
Author one complete voiced quest inside the fixed Red Sands browser game, including dialogue, travel, a gunfight, a cutscene, journal state, and an ending.
xhigh reasoningHeadline result
- Workflow cost
- ¥235.23 estimated API-list-price equivalent (upper bound)
- Wall-clock
- 1h 5m 52.3s wall-clock
- Processed tokens
- 19.32M processed
- Record state
- ledger
Public summary
Qwen3.8 Max xhigh public partial ledger: 1h 5m 52.3s wall-clock, 19.32M processed, and ¥235.23 estimated API-list-price equivalent (upper bound).
Cost basis
- Computed here for comparability: applying the pricing page's generic 10% cache-hit rule (¥1.2/M) to the 18,405,588 cache-read tokens gives ¥36.45. The receipt reports only the ¥235.23 upper bound.
- This row headlines the full-input-rate UPPER BOUND, matching the Stage 1 STARFALL, Fighter Game and Cathedral rows. The Stage 2 Landfall and JRPG rows headline the 10% cache-hit rule instead. Convert before comparing Qwen rows.
- Qwen Code records no per-request cost field; null by construction.
- Pricing source recorded as: official Alibaba Model Studio rate card, Beijing region, resolved live.
- Non-USD upper bound; excluded from USD averages and the cost × time plot.
Visible evidence gaps
- a human playthrough of the quest, start to finish
- blind-evaluation record — the primary instrument for this task
- reconciliation of the root ledger-vs-transcript token gap
- the handback note the prompt asks for
Builder test available
This result is part of a Builder test. Open it for the exact prompt and any released projects, RemakeBench Harness workflows and production skills. Public proof and known evidence gaps stay visible here.
- Red Sands v1 · Builder projects v1