Included
- Model-hard on the models you name, against the pass line you set
- Spread across the specialties you choose
- Question, gold-standard answer, rubric and graded model answers for every case
- Tagged by specialty and type of clinical work
We make dental AI reliable, safe, and effective in clinical settings. Our expert network writes evals to improve AI use with gold-standard answers.
Score on dentist-written case questions, by specialty. Pick a specialty to see how the top models do.
| Model | Case score |
|---|---|
| Dentist's answer | 88% |
| Claude Opus 5.5 | 32% |
| Gemini 4 Argon | 19% |
| GPT-6 Astra | 18% |
| Model | Case score |
|---|---|
| Dentist's answer | 97% |
| Gemini 4 Argon | 29% |
| Claude Opus 5.5 | 25% |
| GPT-6 Astra | 18% |
| Model | Case score |
|---|---|
| Dentist's answer | 93% |
| Claude Opus 5.5 | 24% |
| Gemini 4 Argon | 20% |
| GPT-6 Astra | 16% |
| Model | Case score |
|---|---|
| Dentist's answer | 90% |
| Gemini 4 Argon | 29% |
| Claude Opus 5.5 | 27% |
| GPT-6 Astra | 19% |
| Model | Case score |
|---|---|
| Dentist's answer | 91% |
| Claude Opus 5.5 | 35% |
| Gemini 4 Argon | 27% |
| GPT-6 Astra | 20% |
| Model | Case score |
|---|---|
| Dentist's answer | 91% |
| Claude Opus 5.5 | 24% |
| Gemini 4 Argon | 21% |
| GPT-6 Astra | 16% |
| Model | Case score |
|---|---|
| Dentist's answer | 98% |
| Gemini 4 Argon | 28% |
| Claude Opus 5.5 | 24% |
| GPT-6 Astra | 18% |
Two minutes on what we make, who writes it, and why models still get dental care wrong.
Founder, Frontier Dental · 2 min
Every case is written by a dentist in active practice. These are the people behind the cases.
12practicing dentists
Practicing dentist? Write cases with us
Two ways to start.
Dentists write the cases. We keep only the ones models get wrong.
See our processFrontier Dental owns the cases it produces. Licence terms, including who else may receive a case, are agreed per project.
No. Every case is a composite written from a dentist's experience. No names, dates or details that could identify a real patient.
On the sample case, each answer was graded line by line by the two other models. A line counts in full when both graders agree it is met, and half when only one does.
Yes. You set the bar for your project.
25 cases, delivered in 1 month, spread across specialties.
Five dentist-written eval cases, right in your inbox.