Benchmark Almanac

NPCBench

Claude Sonnet 4.6 23% leads since November 2026
Not in model cards

Models follow patients with nasopharyngeal carcinoma through diagnosis and care. The main patient-level measure requires correct decisions at successive stages.

Made by
Pengkai Wang and colleagues
Released
November 2026
Kind
Medicine
Level
Expert
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
In and out
Text, Images → Text
Interaction
Multi-turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemini 3.1 Pro #40 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
21
Entries in its paper, 13 models we track

Claude Sonnet 4.6 leads with 23% among models released by April 2026 Claude Sonnet 4.6 leads with 23% among models released by April 2026

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper gives no overall score. The headline is strict patient success on CarePath episodes. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude Sonnet 4.6 23%
  2. 2 Gemini-2.5-Pro (snapshot unspecified) 15%
  3. 3 Gemini 3.1 Pro 11%
  4. 4 Hulu-Med-32B 10%
  5. 5 Lingshu-7B 9.00%
  6. 6 Gemini 2.5 Flash Sep 2025 6.00%
  7. 7 Qwen3-VL-30B-A3B-Instruct 6.00%
  8. 8 Qwen 3.6 35B-A3B 5.00%
  9. 9 Qwen3.6 27B 5.00%
  10. 10 Lingshu-32B 2.00%
  11. 11 GPT-4.1 2.00%
  12. 12 Qwen3-VL-8B-Instruct 2.00%
  13. 13 GPT-5.1 2.00%
  14. 14 Qwen3.5-27B 1.00%
  15. 15 Qwen3.5-35B-A3B 1.00%
  16. 16 Qwen3.5-9B 1.00%
  17. 17 HealthGPT-Pro-4B 0%
  18. 18 HealthGPT-Pro-8B 0%
  19. 19 Hulu-Med-4B 0%
  20. 20 Hulu-Med-7B 0%
  21. 21 InternVL3.5-38B 0%

Source: The NPCBench paper

#EntryScoreSettingDateSource
1 Claude Sonnet 4.6 23% Feb 2026 Official leaderboard
2 Gemini-2.5-Pro (snapshot unspecified) 15% Official leaderboard
3 Gemini 3.1 Pro 11% Feb 2026 Official leaderboard
4 Hulu-Med-32B 10% Official leaderboard
5 Lingshu-7B 9.00% Official leaderboard
6 Gemini 2.5 Flash Sep 2025 6.00% Sep 2025 Official leaderboard
7 Qwen3-VL-30B-A3B-Instruct 6.00% Oct 2025 Official leaderboard
8 Qwen 3.6 35B-A3B 5.00% Apr 2026 Official leaderboard
9 Qwen3.6 27B 5.00% Apr 2026 Official leaderboard
10 Lingshu-32B 2.00% Official leaderboard
11 GPT-4.1 2.00% Apr 2025 Official leaderboard
12 Qwen3-VL-8B-Instruct 2.00% Oct 2025 Official leaderboard
13 GPT-5.1 2.00% Nov 2025 Official leaderboard
14 Qwen3.5-27B 1.00% Feb 2026 Official leaderboard
15 Qwen3.5-35B-A3B 1.00% Feb 2026 Official leaderboard
16 Qwen3.5-9B 1.00% Feb 2026 Official leaderboard
17 HealthGPT-Pro-4B 0% Official leaderboard
18 HealthGPT-Pro-8B 0% Official leaderboard
19 Hulu-Med-4B 0% Official leaderboard
20 Hulu-Med-7B 0% Official leaderboard
21 InternVL3.5-38B 0% Aug 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Sonnet 4.6 23% Feb 2026 Official leaderboard
2 Gemini 3.1 Pro 11% Feb 2026 Official leaderboard
3 Gemini 2.5 Flash Sep 2025 6.00% Sep 2025 Official leaderboard
4 Qwen3-VL-30B-A3B-Instruct 6.00% Oct 2025 Official leaderboard
5 Qwen 3.6 35B-A3B 5.00% Apr 2026 Official leaderboard
6 Qwen3.6 27B 5.00% Apr 2026 Official leaderboard
7 GPT-4.1 2.00% Apr 2025 Official leaderboard
8 Qwen3-VL-8B-Instruct 2.00% Oct 2025 Official leaderboard
9 GPT-5.1 2.00% Nov 2025 Official leaderboard
10 Qwen3.5-27B 1.00% Feb 2026 Official leaderboard
11 Qwen3.5-35B-A3B 1.00% Feb 2026 Official leaderboard
12 Qwen3.5-9B 1.00% Feb 2026 Official leaderboard
13 InternVL3.5-38B 0% Aug 2025 Official leaderboard

Rankings by part

We read 13 parts from the benchmark’s paper, each ranking models on one part of the test, such as Guideline recall, Guideline application and Pathology. Gemini 3.1 Pro leads 8 of them, and 5 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Guideline recall

13 models we track have a score here.

  1. 1 Gemini 3.1 Pro ↑1 94%
  2. 2 Claude Sonnet 4.6 ↓1 93.2%
  3. 3 GPT-5.1 ↑6 90.4%
  4. 4 GPT-4.1 ↑3 90%
  5. 5 Qwen3.5-27B ↑5 89.2%
  6. 6 Gemini 2.5 Flash Sep 2025 ↓3 88.4%
  7. 7 InternVL3.5-38B ↑6 88%
  8. 8 Qwen3.6 27B ↓2 87.6%
  9. 9 Qwen3.5-35B-A3B ↑2 86.4%
  10. 10 Qwen3-VL-30B-A3B-Instruct ↓6 86%

Number one over time

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

Anthropic
Benchmark paper 1 leader

Source: The NPCBench paper

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
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Claude Sonnet 4.6 Gemini 2.5 Flash GPT-4.1 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Claude Sonnet 4.6 Gemini 2.5 Flash

Source: Official NPCBench 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for Qwen3.6 27B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (21 entries).
  • Pass Models scored 13 models we track, plus 8 other entries on its leaderboard.
  • Warning Newest score The newest came 7 months ago, in April 2026, for Qwen3.6 27B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.1 Pro (#40).
  • Pass Headroom The best is 23%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 13 rows for models we track and 8 other entries 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