Shareable benchmark table

Runtime, tokens, and recorded cost

Operational aggregates across 35 complete reports per system.

Table 06 · Operational summary

Runtime, tokens, and recorded cost

Active researchReconstructed scoring completeFrozen2 retained non-successesOfficial judge not run
Operational aggregates across 35 complete reports per system.
SystemMedian wallMean tokensTotal tokensMean costTotal cost
GPT-5.6 Sol · High4m 24s99.87k3.5m$0.75 (API-equivalent)$26.08
Grok 4.6 Build · High10m 25s124.16k4.35m$0.41 (provider-reported)$14.31
Table 06Operational aggregates across 35 complete reports per system.

Grok cost is provider-reported; GPT-5.6 Sol cost is API-equivalent. The two price bases are not invoice-equivalent.

BMSbms.kyokasuigetsu.xyz
Static image: https://bms.kyokasuigetsu.xyz/tables/campaign-operations.png