Benchmark Almanac

TREC BioGen 2024

KARL (par. N = 10) 86.7 tops its leaderboard Claude Opus 4.6 79.9 leads the models we track
Not in model cards

Systems retrieve biomedical literature and synthesize evidence into reports that answer a question.

Databricks independently ran the 2024 track for the KARL evaluation and reported results for seven external models.

Kind
Medicine
Level
Expert
Scored
Graded by people
Answers
Open-ended
Tasks
All public
Measures
Capability
Subject
Basic medicine
Language
English
In and out
Text, Documents → Text, Documents
Interaction
Agentic
Environment
Browser
Run by
A host
Occupations
Healthcare practitioners and technical
Strongest model tested
Claude Opus 4.6 #39 on the capability index
Last model added
February 2026
Scores refreshed
November 21st, 2026
16
Entries on its own leaderboard, 10 models we track

KARL (par. N = 10) tops the leaderboard with 86.7 Claude Opus 4.6 leads with 79.9 among models released by February 2026

Best score per entry from the benchmark's own leaderboard, read on November 21st, 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 best model we track is Claude Opus 4.6, at 79.9. This view shows only models we track, each at its best.

The newest model with a score here came out in February 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 KARL (par. N = 10) 86.7
  2. 2 KARL (par. N = 20) 86.7
  3. 3 KARL-TREC 85.0
  4. 4 KARL (par. N = 3) 83.7
  5. 5 KARL 80.2
  6. 6 Claude Opus 4.6 79.9
  7. 7 Claude Sonnet 4.6 77.7
  8. 8 Claude Sonnet 4.5 75.2
  9. 9 Claude Opus 4.5 74.7
  10. 10 Claude Haiku 4.5 72.4
  11. 11 MiniMax-M2.5 69.3
  12. 12 GPT-5 68.3
  13. 13 Qwen3.5 397B-A17B 68.2
  14. 14 KARL-BCP 68.0
  15. 15 GLM-4.5-Air 66.0
  16. 16 GPT-5.2 47.8

Source: Official TREC BioGen 2024 leaderboard

#EntryScoreSettingDateSource
1 KARL (par. N = 10) 86.7 Official leaderboard
2 KARL (par. N = 20) 86.7 Official leaderboard
3 KARL-TREC 85.0 Official leaderboard
4 KARL (par. N = 3) 83.7 Official leaderboard
5 KARL 80.2 Official leaderboard
6 Claude Opus 4.6 79.9 Feb 2026 Official leaderboard
7 Claude Sonnet 4.6 77.7 Feb 2026 Official leaderboard
8 Claude Sonnet 4.5 75.2 Sep 2025 Official leaderboard
9 Claude Opus 4.5 74.7 Nov 2025 Official leaderboard
10 Claude Haiku 4.5 72.4 Oct 2025 Official leaderboard
11 MiniMax-M2.5 69.3 Feb 2026 Official leaderboard
12 GPT-5 68.3 Aug 2025 Official leaderboard
13 Qwen3.5 397B-A17B 68.2 Feb 2026 Official leaderboard
14 KARL-BCP 68.0 Official leaderboard
15 GLM-4.5-Air 66.0 Jul 2025 Official leaderboard
16 GPT-5.2 47.8 Dec 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Opus 4.6 79.9 Feb 2026 Official leaderboard
2 Claude Sonnet 4.6 77.7 Feb 2026 Official leaderboard
3 Claude Sonnet 4.5 75.2 Sep 2025 Official leaderboard
4 Claude Opus 4.5 74.7 Nov 2025 Official leaderboard
5 Claude Haiku 4.5 72.4 Oct 2025 Official leaderboard
6 MiniMax-M2.5 69.3 Feb 2026 Official leaderboard
7 GPT-5 68.3 Aug 2025 Official leaderboard
8 Qwen3.5 397B-A17B 68.2 Feb 2026 Official leaderboard
9 GLM-4.5-Air 66.0 Jul 2025 Official leaderboard
10 GPT-5.2 47.8 Dec 2025 Official leaderboard

4 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.

Z.ai OpenAI Anthropic
Own leaderboard Not tested since Feb 2026

Source: Official TREC BioGen 2024 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
40.0 50.0 60.0 70.0 80.0 2026 40.0 50.0 60.0 70.0 80.0 2026

Source: Official TREC BioGen 2024 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Claude Sonnet 4.6. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (16 entries).
  • Pass Models scored 10 models we track, plus 6 other entries on its leaderboard.
  • Warning Newest score The newest came 9 months ago, in February 2026, for Claude Sonnet 4.6.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 4.6 (#39).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on trec.nist.gov: 10 rows for models we track and 6 other entries at the last read.
  • Pass Last read November 21st, 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