prinzbench
GPT-5.6 Sol Pro 91.0 leads since Jul 2026 Not in model cards
Models research fact patterns under U.S. statutes and regulations and hunt for obscure public information online. 33 private questions run three times give a score out of 99.
25 legal research and 8 search questions. The maker reported gpt-5.6-sol-pro at 91 of 99 on 2026-07-18. Zvi covers it often.
Made by prinz-ai Released Jan 2026 Law Tasks kept private
Last model added Jul 2026
Scores refreshed 2 Nov 2026
- Level
- Expert Specialists with years of training, often over hours of work
- Scored
- Graded by people Human experts grade or compare the outputs.
- Answers
- Open-ended Free-form text, documents, designs or media judged as a whole.
27
Entries on its own leaderboard, 26 models we track
GPT-5.6 Sol Pro leads with 91.0 among models released by Jul 2026
Best score per entry from the benchmark's own leaderboard, read on 2 Nov 2026. 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.
The newest model with a score here came out in Jul 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 10 more
- 91.0 GPT-5.6 Sol Pro
- 82.0 GPT-5.5 Pro
- 79.0 GPT-5.4 Pro
- 74.0 GPT-5.5
- 69.0 GPT-5.4
- 56.0 Claude Fable 5
- 52.0 GPT-5.2
- 52.0 GPT-5.3 Codex
- 50.0
Gemini 3.1 Pro - 47.0 Kimi K3
- 43.0 Grok 4.20
- 42.0 Claude Opus 4.8
- 36.0 Gemini 3 Flash
- 35.0 Gemini 3 Pro
- 35.0 Kimi K2.5
- 31.0
Muse Spark - 30.0 GLM-5.2
- 25.0 qwen-q3-max
- 25.0 Grok 4.1
- 25.0 Claude Opus 4.7
- 23.0 Grok 4
- 23.0
DeepSeek-V4-Pro - 22.0 Kimi K2 Thinking
- 20.0 Claude Sonnet 4.5
- 19.0 Claude Opus 4.6
- 14.0 Claude Opus 4.5
- 8.00 Claude Sonnet 4.6
- 1 GPT-5.6 Sol Pro 91.0
- 2 GPT-5.5 Pro extended 82.0
- 3 GPT-5.4 Pro extended 79.0
- 4 GPT-5.5 thinking heavy 74.0
- 5 GPT-5.4 xhigh 69.0
- 6 Claude Fable 5 max 56.0
- 7 GPT-5.2 thinking extended 52.0
- 8 GPT-5.3 Codex high 52.0
- 9 Gemini 3.1 Pro 50.0
- 10 Kimi K3 47.0
- 11 Grok 4.20 43.0
- 12 Claude Opus 4.8 max 42.0
- 13 Gemini 3 Flash 36.0
- 14 Gemini 3 Pro 35.0
- 15 Kimi K2.5 thinking 35.0
- 16 Muse Spark 31.0
- 17 GLM-5.2 30.0
- 18 qwen-q3-max 25.0
- 19 Grok 4.1 thinking 25.0
- 20 Claude Opus 4.7 25.0
- 21 Grok 4 23.0
- 22 DeepSeek-V4-Pro 23.0
- 23 Kimi K2 Thinking 22.0
- 24 Claude Sonnet 4.5 20.0
- 25 Claude Opus 4.6 19.0
- 26 Claude Opus 4.5 14.0
- 27 Claude Sonnet 4.6 8.00
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.6 Sol Pro | 91.0 | Jul 2026 | Official leaderboard | |
| 2 | GPT-5.5 Pro | 82.0 | extended | Apr 2026 | Official leaderboard |
| 3 | GPT-5.4 Pro | 79.0 | extended | Mar 2026 | Official leaderboard |
| 4 | GPT-5.5 | 74.0 | thinking heavy | Apr 2026 | Official leaderboard |
| 5 | GPT-5.4 | 69.0 | xhigh | Mar 2026 | Official leaderboard |
| 6 | Claude Fable 5 | 56.0 | max | Jun 2026 | Official leaderboard |
| 7 | GPT-5.2 | 52.0 | thinking extended | Dec 2025 | Official leaderboard |
| 8 | GPT-5.3 Codex | 52.0 | high | Feb 2026 | Official leaderboard |
| 9 | Gemini 3.1 Pro | 50.0 | Feb 2026 | Official leaderboard | |
| 10 | Kimi K3 | 47.0 | Jul 2026 | Official leaderboard | |
| 11 | Grok 4.20 | 43.0 | Feb 2026 | Official leaderboard | |
| 12 | Claude Opus 4.8 | 42.0 | max | May 2026 | Official leaderboard |
| 13 | Gemini 3 Flash | 36.0 | Dec 2025 | Official leaderboard | |
| 14 | Gemini 3 Pro | 35.0 | Nov 2025 | Official leaderboard | |
| 15 | Kimi K2.5 | 35.0 | thinking | Jan 2026 | Official leaderboard |
| 16 | Muse Spark | 31.0 | Apr 2026 | Official leaderboard | |
| 17 | GLM-5.2 | 30.0 | Jun 2026 | Official leaderboard | |
| 18 | qwen-q3-max | 25.0 | Official leaderboard | ||
| 19 | Grok 4.1 | 25.0 | thinking | Nov 2025 | Official leaderboard |
| 20 | Claude Opus 4.7 | 25.0 | Apr 2026 | Official leaderboard | |
| 21 | Grok 4 | 23.0 | Jul 2025 | Official leaderboard | |
| 22 | DeepSeek-V4-Pro | 23.0 | Apr 2026 | Official leaderboard | |
| 23 | Kimi K2 Thinking | 22.0 | Nov 2025 | Official leaderboard | |
| 24 | Claude Sonnet 4.5 | 20.0 | Sep 2025 | Official leaderboard | |
| 25 | Claude Opus 4.6 | 19.0 | Feb 2026 | Official leaderboard | |
| 26 | Claude Opus 4.5 | 14.0 | Nov 2025 | Official leaderboard | |
| 27 | Claude Sonnet 4.6 | 8.00 | Feb 2026 | Official leaderboard |
4 models have held number one since its release in Jan 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
OpenAI
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
On the benchmark’s own leaderboard
No lab model card reports it yet
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.
More in Law, finance and medicine
- 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%
- 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 In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
- KBL In use Models answer Korean legal knowledge and reasoning tasks and Korean bar exam questions. It has 7 knowledge tasks, 4 reasoning tasks and 2,510 bar exam items.
Solar Pro 4 77.5% - 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. Solar Pro 4 93.2%
- PRBench In use Model answers 1,100 expert-written law and finance questions. LLM judge grades against weighted expert rubrics of 10-30 criteria each. Muse Spark 1.3 61.6%