Benchmark Almanac

Periapical Lesions Bench

Gemini 4 Neon 18.7% leads since November 2026
Not in model cards

Agents locate and grade lesions around tooth roots in panoramic dental X-rays. A perfect report matches every lesion and grade without adding extra boxes.

On November 13, 2026, Gemini 4 Carbon led seven models with 15.7% pass@1 on 100 films. GPT-6.2 Astra led on mean reward at 0.431. The PanoRead board uses 200 films and one attempt per film. GPT-6 Astra led that board on October 4, 2026, with 0.224 mean reward.

Made by
Dylan Tirandaz
Released
September 2026
Kind
Medicine
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images → Actions, Text
Interaction
Agentic
Environment
Desktop apps
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
8
Models scored on its own leaderboard

Gemini 4 Neon leads with 18.7%

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.

  1. 1 Gemini 4 Neon 18.7%
  2. 2 Gemini 4 Carbon 15.7%
  3. 3 GPT-6.2 Astra 12%
  4. 4 Claude Opus 5.5 8.30%
  5. 5 GPT-6.1 Sol 6.30%
  6. 6 Kimi K3.1 Pro 2.30%
  7. 7 GLM-5.3-Flash 1.00%
  8. 8 DeepSeek V4.1 Flash 0.30%

Source: Official Periapical Lesions Bench leaderboard

#ModelScoreSettingDateSource
1 Gemini 4 Neon 18.7% Nov 2026 Official leaderboard
2 Gemini 4 Carbon 15.7% Oct 2026 Official leaderboard
3 GPT-6.2 Astra 12% Nov 2026 Official leaderboard
4 Claude Opus 5.5 8.30% Sep 2026 Official leaderboard
5 GPT-6.1 Sol 6.30% Sep 2026 Official leaderboard
6 Kimi K3.1 Pro 2.30% Oct 2026 Official leaderboard
7 GLM-5.3-Flash 1.00% Aug 2026 Official leaderboard
8 DeepSeek V4.1 Flash 0.30% Sep 2026 Official leaderboard

The best score, 18.7%, costs $1.97 a task

Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.

Z.ai DeepSeek OpenAI Google DeepMind Cost frontier One model at each effort level
0% 5% 10% 15% 20% $0.01 $0.10 $1.00 $10 Cost per task, log scale Gemini 4 Neon Gemini 4 Carbon GPT-6.2 Astra GPT-6.1 Sol GLM-5.3-Flash DeepSeek V4.1 Flash 0% 5% 10% 15% 20% $0.01 $0.10 $1.00 $10 Cost per task, log scale Gemini 4 Neon Gemini 4 Carbon

Source: Official Periapical Lesions Bench leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 4 Neon 18.7% $1.97
Gemini 4 Carbon 15.7% $1.77
GPT-6.2 Astra 12% $0.48
Claude Opus 5.5 8.30% $1.58
GPT-6.1 Sol 6.30% $0.15
Kimi K3.1 Pro 2.30% $0.37
GLM-5.3-Flash 1.00% $0.09
DeepSeek V4.1 Flash 0.30% $0.05

4 models have held number one on the official leaderboard since its release in September 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Z.ai Anthropic Google DeepMind
Own leaderboard 4 leaders

Source: Official Periapical Lesions Bench leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) Gemini 4 Neon Claude Opus 5.5 GLM-5.3-Flash 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) Gemini 4 Neon Claude Opus 5.5

Source: Official Periapical Lesions Bench leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Law, finance and medicine

All 637 in Law, finance and medicine
  • MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy. Qwen 3.8 Max 80.4%
  • HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
  • HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy. Ling-3.1-flash 81.7%
  • KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
  • Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%

Data audit

All 11 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 5 pass
  • Pass Independent runs Scores from the official leaderboard (8 entries).
  • Pass Models scored 8 models we track.
  • Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
  • Pass Frontier models 4 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Pass Headroom The best is 18.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on dylantirandaz.com: 8 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check