ModelDial / Measured model comparisons
GPT-6.1 Sol vs GPT-6 Sol
Compare the two Sol generations at their highest-scoring tested overall efforts, including whether score changes come with time or cost changes.
See the axis results below. Missing overall scores or different frontend scoring versions prevent an overall difference claim.
Overall: backend 40% · frontend 30% · reasoning 30%Scores, time and cost
| Metric | GPT-6.1 Solmax | GPT-6 Solxhigh |
|---|---|---|
| Overall | 93.8 | 82.7 |
| Backend & testing | 95 | 80 |
| Frontend & interaction | 96 | 94 |
| Knowledge & reasoning | 90 | 75 |
| Three-axis time | 62m 9s | 39m 22s |
| Three-axis cost | $1.69 | $1.10 |
Each column uses three axes from one configuration, without mixing efforts. Costs exclude subscriptions and failed retries; missing values are shown as “—”.
Frontend scoring versions may differ. Open the full comparison to inspect the grading basis.
How to read these results
What should I check before switching from GPT-6 Sol to GPT-6.1 Sol?
Check the capability your task needs first, then compare time and cost at the relevant efforts. This page selects the highest overall score for each model. If you use a fixed effort, select it in the full comparison so model changes and effort changes remain distinguishable.
This page uses ModelDial’s published evaluations, not official vendor rankings. Model detail pages provide the axis results, scoring versions and tested configurations.