Human baseline
i
02:00:00
Reference · Level 10
02:00:00
Reference · Level 10
Tokens and cost not yet verified.
17:18:51
Finished · Level 10
26Tokens and cost not yet verified.
23:29:07
Finished · Level 10
Tokens and cost not yet verified.
26:04:53
Finished · Level 10
35Tokens and cost not yet verified.
02:13:55
Stopped · Level 4
09:06:00
Running · Level 4
04:37:21
Stopped · Level 3
1· 2,097 requests
| Tokens | Count | USD / 1M | Cost |
|---|---|---|---|
| Fresh input | 7,543,035 | $0.75 | $5.66 |
| Cached input | 267,316,771 | $0.075 | $20.05 |
| Output | 341,061 | $3.75 | $1.28 |
| Total | 275,200,867 | $26.99 |
Discounts included: 50% introductory pricing through 31 Dec 2026; cached input is a further 90% cheaper than fresh input.
Standard interactive API equivalent, not an invoice or subscription charge. No Batch or Flex discount applied. Tokens include repeated cached context; output includes reported thinking without counting it twice.
Rates checked 2026-10-26: Google pricing · Introductory & caching discounts. Other runs’ usage is not yet verified.
· 691 requests
| Tokens | Count | USD / 1M | Cost |
|---|---|---|---|
| Fresh input | 2,042,186 | $2.00 | $4.08 |
| Cached input | 90,470,691 | $0.10 | $9.05 |
| Output | 58,747 | $10.00 | $0.59 |
| Total | 92,571,624 | $13.72 |
Discounts included: 50% introductory pricing through 30 Nov 2026; cached input is a further 95% cheaper than fresh input.
Standard interactive API equivalent, not an invoice or subscription charge. No Batch or Flex discount applied. Tokens include repeated cached context; output includes reported thinking without counting it twice.
Rates checked 2026-10-25: Google pricing · Introductory & caching discounts. Other runs’ usage is not yet verified.
How to read it
Orc warrior. Level 1 → 10. /played is time logged in, including thinking and troubleshooting. Level 10 time is the score; unfinished runs show their last level.
Newer clean v2 runs use WotLK 3.3.5a AzerothCore x1. Legacy runs had different controls.
Conditions differ. The human baseline is a reference, not a controlled trial.
Source video
Two GPT-6.1 Sol attempts, from the first step to level 10.
Open on YouTube