Chess with an engine
loads medium
Overall69
Score5.5/8
Tech90
UX / UI33 / 90
Time2 min 56 s
Cost$0.28
Compare →
GPT-6 Sol
OpenAIReleased 22/09/2026
Overall66/100#11 of 27
Tech92/100#10 of 27
UX / UI54 / 64#15 of 27
Score26.5/40 (33.5)
Total cost$1.25
Average time2 min 48 s#12 of 27
Total tokens1,319,789
API price: $2 / $10 per 1M input / output tokens · as of 15/10/2026 source ↗
What is GPT-6 Sol good for?
OpenAI's model built for complex coding and agentic workflows.
Description based on the maker · source ↗
- Maker
- OpenAI
- Released
- 22/09/2026
- Model weights
- closed, the model is available through the API only
- Reasoning levels
- low medium high
How did GPT-6 Sol do in the AI Lab tests?
- 4 of 5 outputs load without errors; 1 more ran after one fix by the same model.
- Its best technical result was on the “Solar System” prompt (100/100).
- Did not work (original output): GTA-style game.
- In the objective perft chess test it passed 2 of 6 positions.
- Average time per prompt 2 min 48 s#12 of 27
- Average cost per prompt $0.25#17 of 27
- Overall score 66/100#11 of 27
- Technical health 92/100#10 of 27
- Automatic UX rating 54/100#15 of 27
Outputs
All outputs of GPT-6 Sol on the 5 prompts every other model got as well.
Solar System
loads medium
Overall83
Score6.5/8
Tech100
UX / UI79 / 80
Time2 min 27 s
Cost$0.26
Compare →
GTA-style game
broken medium fixed version →
Overall0
Score0/8
Tech70
UX / UI0 / 0
Time3 min 11 s
Cost$0.29
Compare →
Monitor test
loads medium
Overall96
Score8/8
Tech100
UX / UI90 / 80
Time2 min 15 s
Cost$0.18
Compare →
New York & Central Park map
loads medium
Overall80
Score6.5/8
Tech100
UX / UI70 / 70
Time3 min 13 s
Cost$0.24
Compare →
Cost = tokens priced with the API price list valid on the date shown next to the cost (prices change). ≈ = estimate (only the total token count is known), no ≈ = exact from input / output tokens. “Loads” = the page opens without errors; technical health also checks behaviour during use – judge the quality in the viewer.