Benchmark Almanac

PharmaBench (micro1)

Kimi K3.1 Pro 27.5% leads since Oct 2026 Not in model cards

Agents write clinical trial documentation inside 8 synthetic drug programs, across 120 tasks written by biostatisticians, medical writers, safety physicians and regulatory staff.

Penalties subtract for content an expert would have to remove. Full marks were rare, 17.2% of runs for the best model. Not the 2024 PharmaBench of ADMET data.

Made by micro1 Released Oct 2026 Medicine Last model added Oct 2026 Scores refreshed 2 Nov 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Graded by a model A language model scores each response, usually against a rubric.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
Paper Leaderboard
12
Models scored on its own leaderboard

Kimi K3.1 Pro leads with 27.5%

Best score per model from the benchmark's own leaderboard, read on 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best.

  1. 27.5 Kimi K3.1 Pro
  2. 17.2 Muse Wonder
  3. 16.1 GPT-6 Astra
  4. 13.9 Muse Spark 1.3
  5. 12.8 Claude Fable 5.1
  6. 12.5 Grok 4.6
  7. 11.4 Claude Sonnet 5.5
  8. 10.8 Claude Opus 5.5
  9. 9.70 Muse Spark 1.1
  10. 9.70 Claude Opus 5
  11. 8.60 Gemini 4 Argon
  12. 5.60 Gemini 3.8 Flash
  1. 1 Kimi K3.1 Pro 27.5%
  2. 2 Muse Wonder 17.2%
  3. 3 GPT-6 Astra 16.1%
  4. 4 Muse Spark 1.3 13.9%
  5. 5 Claude Fable 5.1 12.8%
  6. 6 Grok 4.6 12.5%
  7. 7 Claude Sonnet 5.5 11.4%
  8. 8 Claude Opus 5.5 10.8%
  9. 9 Muse Spark 1.1 9.70%
  10. 10 Claude Opus 5 9.70%
  11. 11 Gemini 4 Argon 8.60%
  12. 12 Gemini 3.8 Flash 5.60%
#ModelScoreSettingDateSource
1 Kimi K3.1 Pro 27.5% Oct 2026 Official leaderboard
2 Muse Wonder 17.2% Oct 2026 Official leaderboard
3 GPT-6 Astra 16.1% Sep 2026 Official leaderboard
4 Muse Spark 1.3 13.9% Sep 2026 Official leaderboard
5 Claude Fable 5.1 12.8% Sep 2026 Official leaderboard
6 Grok 4.6 12.5% Aug 2026 Official leaderboard
7 Claude Sonnet 5.5 11.4% Sep 2026 Official leaderboard
8 Claude Opus 5.5 10.8% Sep 2026 Official leaderboard
9 Muse Spark 1.1 9.70% Jul 2026 Official leaderboard
10 Claude Opus 5 9.70% Jul 2026 Official leaderboard
11 Gemini 4 Argon 8.60% Sep 2026 Official leaderboard
12 Gemini 3.8 Flash 5.60% Sep 2026 Official leaderboard

3 models have held number one since its release in Oct 2026

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

OpenAI Meta Moonshot AI
Own leaderboard 3 leaders
Kimi K3.1 Pro 27.5%

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 Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) Kimi K3.1 Pro Muse Wonder GPT-6 Astra Muse Spark 1.1 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) Kimi K3.1 Pro Muse Wonder

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

All 140 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%