Benchmark Almanac

VetEval

Claude Opus 5.5 76.8 leads since September 2026
Not in model cards

Models answer examination questions written by licensed veterinarians. A fixed answer key measures clinical knowledge, and safety penalties reduce the overall index.

The scored items remain private. VetEval ranks 29 models and publishes separate practice questions.

Made by
viggoVet
Kind
Medicine
Level
University
Scored
Checked automatically
Answers
Multiple choice
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Written by experts
Refresh
One fixed set
Run by
A host
Occupations
Healthcare practitioners and technical
In views
Content safety
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
29
Models scored on its own leaderboard

Claude Opus 5.5 leads with 76.8

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.

Source: Official VetEval leaderboard

#ModelScore ±SettingSubmitted byReleasedRun onLinkSource
1 Claude Opus 5.5 76.8 ±2 vendor_infra VetEval Sep 2026 Sep 2026 veteval.ai Official leaderboard
2 Kimi K3.1 Pro 71.6 69.6 to 73.4 vendor_infra VetEval Oct 2026 Oct 2026 veteval.ai Official leaderboard
3 GPT-6.2 Astra 71.4 69.3 to 73.6 vendor_infra VetEval Nov 2026 Nov 2026 veteval.ai Official leaderboard
4 GPT-5.6 Sol 69.8 67.7 to 71.8 vendor_infra VetEval Jul 2026 Sep 2026 veteval.ai Official leaderboard
5 Gemini 3.7 Flash 66.4 ±2.1 vendor_infra VetEval Aug 2026 Aug 2026 veteval.ai Official leaderboard
6 Gemini 4 Carbon 66.2 64.3 to 68.1 vendor_infra VetEval Oct 2026 Nov 2026 veteval.ai Official leaderboard
7 Gemini 4 Argon 64.5 ±2.1 vendor_infra VetEval Sep 2026 Oct 2026 veteval.ai Official leaderboard
8 Claude Opus 5 63.9 62.2 to 65.7 vendor_infra VetEval Jul 2026 Sep 2026 veteval.ai Official leaderboard
9 GPT-6 Sol 62.4 ±2.1 vendor_infra VetEval Sep 2026 Sep 2026 veteval.ai Official leaderboard
10 Gemini 4 Hydrogen 62.3 60.4 to 64.4 vendor_infra VetEval Nov 2026 Nov 2026 veteval.ai Official leaderboard
11 GPT-6.2 Sol 61.8 59.8 to 63.9 vendor_infra VetEval Nov 2026 Nov 2026 veteval.ai Official leaderboard
12 Muse Wonder 58.3 ±2 vendor_infra VetEval Oct 2026 Nov 2026 veteval.ai Official leaderboard
13 GPT-5.6 Terra 57.6 ±1.9 vendor_infra VetEval Jul 2026 Aug 2026 veteval.ai Official leaderboard
14 GPT-6 Luna 55.7 53.7 to 57.6 vendor_infra VetEval Sep 2026 Sep 2026 veteval.ai Official leaderboard
15 Kimi K3 55.5 53.9 to 57 vendor_infra VetEval Jul 2026 Aug 2026 veteval.ai Official leaderboard
16 Gemini 3.8 Flash 55.1 ±2 vendor_infra VetEval Sep 2026 Sep 2026 veteval.ai Official leaderboard
17 GLM-5.3 54.1 ±1.9 vendor_infra VetEval Aug 2026 Aug 2026 veteval.ai Official leaderboard
18 Grok 4.6 54.0 52.2 to 56 vendor_infra VetEval Aug 2026 Aug 2026 veteval.ai Official leaderboard
19 Grok 4.7 48.6 46.7 to 50.5 vendor_infra VetEval Sep 2026 Nov 2026 veteval.ai Official leaderboard
20 GLM-5.3-Flash 48.5 46.9 to 50.2 vendor_infra VetEval Aug 2026 Sep 2026 veteval.ai Official leaderboard
21 Claude Sonnet 5 46.4 ±2 vendor_infra VetEval Jun 2026 Aug 2026 veteval.ai Official leaderboard
22 Mistral Large 4 45.7 ±1.7 vendor_infra VetEval Oct 2026 Nov 2026 veteval.ai Official leaderboard
23 MiniMax-M3 38.4 ±1.6 vendor_infra VetEval Jun 2026 Aug 2026 veteval.ai Official leaderboard
24 Llama 4 Maverick 38.1 ±1.7 vendor_infra VetEval Apr 2025 Aug 2026 veteval.ai Official leaderboard
25 Claude Haiku 5.5 32.7 ±1.6 vendor_infra VetEval Oct 2026 Oct 2026 veteval.ai Official leaderboard
26 Tencent Hy4 31.5 30.1 to 33 vendor_infra VetEval Nov 2026 Nov 2026 veteval.ai Official leaderboard
27 Qwen3.8-2.4T-A95B 25.3 24.3 to 26.3 vendor_infra VetEval Aug 2026 Aug 2026 veteval.ai Official leaderboard
28 Nemotron 3.5 Lightning 30B-A3B 3.7 ±0.5 vendor_infra VetEval Aug 2026 Aug 2026 veteval.ai Official leaderboard
29 Granite 4.2 8B 3.4 ±0.3 vendor_infra VetEval Aug 2026 Sep 2026 veteval.ai Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Clinical Diagnosis & Differential, Pharmacology & Dosing and Species-Specific Knowledge. Claude Opus 5.5 leads 3 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Clinical Diagnosis & Differential

29 models we track have a score here.

  1. 1 Claude Opus 5.5 90.5
  2. 2 Kimi K3 ↑13 87.7
  3. 3 GPT-6.2 Astra 87.3
  4. 4 GPT-6.2 Sol ↑7 86.8
  5. 5 GPT-5.6 Sol ↓1 86.7
  6. 6 Kimi K3.1 Pro ↓4 86.7
  7. 7 Gemini 3.7 Flash ↓2 86.6
  8. 8 Gemini 4 Hydrogen ↑2 86.3
  9. 9 Claude Opus 5 ↓1 86.2
  10. 10 Gemini 4 Carbon ↓4 84.9

The best score, 76.8, costs $0.01 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.

NVIDIA Google DeepMind IBM Anthropic OpenAI Moonshot AI Cost frontier One model at each effort level
0.0 20.0 40.0 60.0 80.0 $0.00001 $0.0001 $0.001 $0.01 $0.10 Cost per task, log scale Claude Opus 5.5, vendor_infra Kimi K3.1 Pro, vendor_infra Gemini 3.7 Flash, vendor_infra GPT-6 Sol, vendor_infra GPT-6 Luna, vendor_infra Nemotron 3.5 Lightning 30B-A3B, vendor_infra 0.0 20.0 40.0 60.0 80.0 $0.00001 $0.0001 $0.001 $0.01 $0.10 Cost per task, log scale Claude Opus 5.5, vendor_infra Kimi K3.1 Pro, vendor_infra

Source: Official VetEval leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Opus 5.5 vendor_infra 76.8 $0.01
Kimi K3.1 Pro vendor_infra 71.6 $0.01
GPT-6.2 Astra vendor_infra 71.4 $0.007
GPT-5.6 Sol vendor_infra 69.8 $0.009
Gemini 3.7 Flash vendor_infra 66.4 $0.007
Gemini 4 Carbon vendor_infra 66.2 $0.004
Gemini 4 Argon vendor_infra 64.5 $0.004
Claude Opus 5 vendor_infra 63.9 $0.008
GPT-6 Sol vendor_infra 62.4 $0.003
Gemini 4 Hydrogen vendor_infra 62.3 $0.01
GPT-6.2 Sol vendor_infra 61.8 $0.002
Muse Wonder vendor_infra 58.3 $0.01
GPT-5.6 Terra vendor_infra 57.6 $0.005
GPT-6 Luna vendor_infra 55.7 $0.0003
Kimi K3 vendor_infra 55.5 $0.01
Gemini 3.8 Flash vendor_infra 55.1 $0.008
GLM-5.3 vendor_infra 54.1 $0.0008
Grok 4.6 vendor_infra 54.0 $0.008
Grok 4.7 vendor_infra 48.6 $0.02
GLM-5.3-Flash vendor_infra 48.5
Claude Sonnet 5 vendor_infra 46.4 $0.002
Mistral Large 4 vendor_infra 45.7 $0.02
MiniMax-M3 vendor_infra 38.4 $0.002
Llama 4 Maverick vendor_infra 38.1 $0.001
Claude Haiku 5.5 vendor_infra 32.7 $0.0006
Tencent Hy4 vendor_infra 31.5 $0.04
Qwen3.8-2.4T-A95B vendor_infra 25.3 $0.02
Nemotron 3.5 Lightning 30B-A3B vendor_infra 3.7 $0.0003
Granite 4.2 8B vendor_infra 3.4 $0.00004

5 models have held number one on the official leaderboard since 2023

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

Meta MiniMax Anthropic OpenAI
Own leaderboard 5 leaders

Source: Official VetEval 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
0 20 40 60 80 2025 2026 2027 0 20 40 60 80 2025 2026 2027

Source: Official VetEval 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 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (29 entries).
  • Pass Models scored 29 models we track.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Pass Frontier models 4 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Note Headroom Scored as an index, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on veteval.ai: 29 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