Benchmark Almanac

TxBench-Preclinical-Pharmacology

Claude Opus 5.5 65.3% leads since Sep 2026 Not in model cards

Agents analyze preclinical pharmacology data and make development decisions. A grader checks each decision.

Part of LatchBio's BioBench suite on benchmarks.bio. The V1 set has 100 evaluations. Release month is the GitHub repo creation month.

Made by LatchBio Released Jun 2026 Genomics and life sciences Public tasks plus a private test set Last model added Oct 2026 Scores refreshed 2 Nov 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
latchbio/txbench-pp 8 stars Leaderboard benchmarks.bio
28
Models scored on its own leaderboard

Claude Opus 5.5 leads with 65.3%

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.

28 models. Scroll sideways for the rest.

  1. 65.3 Claude Opus 5.5
  2. 64.7 Gemini 4 Argon
  3. 64.3 Claude Sonnet 5.5
  4. 63.0 Claude Opus 5
  5. 63.0 GPT-6 Sol
  6. 61.7 Grok 4.6
  7. 61.3 GPT-5.6 Sol
  8. 59.7 Claude Fable 5.1
  9. 59.7 GPT-6.1 Sol
  10. 59.3 GPT-6 Astra
  11. 57.7 Claude Opus 4.8
  12. 55.7 GPT-6.1 Terra
  13. 55.3 GPT-5.5
  14. 55.3 Gemini 3.7 Flash
  15. 54.7 Gemini 3.8 Flash
  16. 54.7 Grok 4.7
  17. 49.3 Claude Sonnet 5
  18. 46.7 DeepSeek V4.1 Flash
  19. 45.3 Claude Opus 4.7
  20. 44.3 GPT-6 Luna
  21. 43.7 GPT-5.6 Luna
  22. 42.3 Mistral Large 4
  23. 39.7 Kimi K3
  24. 39.7 Beam 1.0
  25. 35.0 GLM-5.3
  26. 34.3 Claude Sonnet 4.6
  27. 22.7 Nemotron 3 Ultra
  28. 10.3 Nemotron 3.5 Lightning 30B-A3B
  1. 1 Claude Opus 5.5 LSVP 65.3%
  2. 2 Gemini 4 Argon 64.7%
  3. 3 Claude Sonnet 5.5 64.3%
  4. 4 Claude Opus 5 63%
  5. 5 GPT-6 Sol 63%
  6. 6 Grok 4.6 61.7%
  7. 7 GPT-5.6 Sol 61.3%
  8. 8 Claude Fable 5.1 LSVP 59.7%
  9. 9 GPT-6.1 Sol 59.7%
  10. 10 GPT-6 Astra 59.3%
  11. 11 Claude Opus 4.8 57.7%
  12. 12 GPT-6.1 Terra 55.7%
  13. 13 GPT-5.5 55.3%
  14. 14 Gemini 3.7 Flash 55.3%
  15. 15 Gemini 3.8 Flash 54.7%
  16. 16 Grok 4.7 54.7%
  17. 17 Claude Sonnet 5 49.3%
  18. 18 DeepSeek V4.1 Flash 46.7%
  19. 19 Claude Opus 4.7 45.3%
  20. 20 GPT-6 Luna 44.3%
  21. 21 GPT-5.6 Luna 43.7%
  22. 22 Mistral Large 4 42.3%
  23. 23 Kimi K3 39.7%
  24. 24 Beam 1.0 39.7%
  25. 25 GLM-5.3 35%
  26. 26 Claude Sonnet 4.6 34.3%
  27. 27 Nemotron 3 Ultra 22.7%
  28. 28 Nemotron 3.5 Lightning 30B-A3B 10.3%
#ModelScoreSettingDateSource
1 Claude Opus 5.5 65.3% LSVP, Pi Sep 2026 Official leaderboard
2 Gemini 4 Argon 64.7% Pi Sep 2026 Official leaderboard
3 Claude Sonnet 5.5 64.3% Pi Sep 2026 Official leaderboard
4 Claude Opus 5 63% Pi Jul 2026 Official leaderboard
5 GPT-6 Sol 63% Pi Sep 2026 Official leaderboard
6 Grok 4.6 61.7% Pi Aug 2026 Official leaderboard
7 GPT-5.6 Sol 61.3% Pi Jul 2026 Official leaderboard
8 Claude Fable 5.1 59.7% LSVP, Pi Sep 2026 Official leaderboard
9 GPT-6.1 Sol 59.7% Pi Sep 2026 Official leaderboard
10 GPT-6 Astra 59.3% Pi Sep 2026 Official leaderboard
11 Claude Opus 4.8 57.7% Pi May 2026 Official leaderboard
12 GPT-6.1 Terra 55.7% Pi Oct 2026 Official leaderboard
13 GPT-5.5 55.3% Pi Apr 2026 Official leaderboard
14 Gemini 3.7 Flash 55.3% Pi Aug 2026 Official leaderboard
15 Gemini 3.8 Flash 54.7% Pi Sep 2026 Official leaderboard
16 Grok 4.7 54.7% Pi Sep 2026 Official leaderboard
17 Claude Sonnet 5 49.3% Pi Jun 2026 Official leaderboard
18 DeepSeek V4.1 Flash 46.7% Pi Sep 2026 Official leaderboard
19 Claude Opus 4.7 45.3% Pi Apr 2026 Official leaderboard
20 GPT-6 Luna 44.3% Pi Sep 2026 Official leaderboard
21 GPT-5.6 Luna 43.7% Codex Jul 2026 Official leaderboard
22 Mistral Large 4 42.3% Pi Oct 2026 Official leaderboard
23 Kimi K3 39.7% Pi Jul 2026 Official leaderboard
24 Beam 1.0 39.7% Pi Oct 2026 Official leaderboard
25 GLM-5.3 35% Pi Aug 2026 Official leaderboard
26 Claude Sonnet 4.6 34.3% Pi Feb 2026 Official leaderboard
27 Nemotron 3 Ultra 22.7% Pi Jun 2026 Official leaderboard
28 Nemotron 3.5 Lightning 30B-A3B 10.3% Pi Aug 2026 Official leaderboard

4 models have held number one since its release in Jun 2026

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

Anthropic OpenAI
Own leaderboard 4 leaders
Claude Opus 5.5 65.3%

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% 2026 2027 Near the ceiling (85%) Claude Opus 5.5 GPT-5.6 Sol GPT-5.5 Claude Opus 4.7 Claude Sonnet 4.6 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Opus 5.5 Claude Opus 4.8

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 Biology and chemistry

All 96 in Biology and chemistry
  • SUPERChem In use Models solve 500 expert-written chemistry reasoning problems, offered in multimodal and text-only forms. Expert solution paths also allow a reasoning path score. Tencent Hy4 preview 66.4%
  • GeneBench Pro In use Agents analyze messy simulated genomics and biology datasets to estimate a decision-relevant quantity across 129 problems. Scored automatically against known answers. GPT-6 Astra 37.1%
  • LifeSciBench In use Models complete 750 expert-written life science research tasks, often with attached figures, tables, or sequence files. Graded against expert rubrics as score and pass rate. GPT-6 Astra 60.3%
  • MolecularIQ In use Models answer questions about molecular graphs, such as counting, indexing and constrained generation, and a symbolic checker verifies each answer. Intern-S2-397B 62.4%
  • TOMG-Bench In use Models generate molecules from natural language requests in three tasks: molecule editing, molecule optimization and customized generation. Each output is checked for validity and for meeting the requested properties. Intern-S2-397B 66.8%
  • BABE Fading Models reason from experimental results in peer-reviewed biology papers to draw conclusions, with causal and cross-scale inference. Seed2.0 Lite (0428) 57.9%