Benchmark Almanac

TxBench-Oligo-Discovery

Claude Opus 5.5 62.2% leads since September 2026
Not in model cards

Agents make oligonucleotide discovery decisions, such as ASO and siRNA design, from real data. A grader checks each decision.

Part of LatchBio's BioBench suite on benchmarks.bio. The V1 set has 113 evaluations. Release month is the GitHub repo creation month.

Made by
LatchBio
Released
September 2026
Kind
Genomics and life sciences
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Public, plus a private test set
Strongest model tested
Gemini 4 Carbon #2 on the capability index
Last model added
November 2026
Scores refreshed
November 13th, 2026
30
Models scored on its own leaderboard

Claude Opus 5.5 leads with 62.2%

Best score per model from the benchmark's own leaderboard, read on November 13th, 2026. Where a model ran with several settings or harnesses, we keep its best.

30 models. Scroll sideways for the rest.

  1. 62.2 Claude Opus 5.5
  2. 55.8 Gemini 4 Carbon
  3. 54.9 Claude Fable 5.1
  4. 54.3 Gemini 4 Hydrogen
  5. 52.5 GPT-6.1 Sol
  6. 51.9 GPT-6 Astra
  7. 51.9 Claude Sonnet 5.5
  8. 51.0 Gemini 4 Argon
  9. 48.4 Claude Opus 5
  10. 46.0 Grok 4.6
  11. 43.1 GPT-6 Sol
  12. 41.9 Grok 4.7
  13. 37.8 Gemini 3.7 Flash
  14. 37.5 Claude Opus 4.8
  15. 37.2 GPT-5.6 Sol
  16. 36.6 Claude Sonnet 5
  17. 35.1 Gemini 3.8 Flash
  18. 34.8 DeepSeek V4.1 Flash
  19. 33.9 GPT-5.5
  20. 33.6 GPT-6 Luna
  21. 31.9 Claude Opus 4.7
  22. 31.9 GLM-5.3
  23. 31.3 Kimi K3
  24. 30.4 GPT-6.1 Terra
  25. 30.1 Claude Sonnet 4.6
  26. 27.4 Beam 1.0
  27. 27.1 Mistral Large 4
  28. 25.1 Nemotron 3 Ultra
  29. 23.0 GPT-5.6 Luna
  30. 8.55 Nemotron 3.5 Lightning 30B-A3B
  1. 1 Claude Opus 5.5 LSVP 62.2%
  2. 2 Gemini 4 Carbon 55.8%
  3. 3 Claude Fable 5.1 LSVP 54.9%
  4. 4 Gemini 4 Hydrogen 54.3%
  5. 5 GPT-6.1 Sol 52.5%
  6. 6 GPT-6 Astra 51.9%
  7. 7 Claude Sonnet 5.5 51.9%
  8. 8 Gemini 4 Argon 51%
  9. 9 Claude Opus 5 48.4%
  10. 10 Grok 4.6 46%
  11. 11 GPT-6 Sol 43.1%
  12. 12 Grok 4.7 41.9%
  13. 13 Gemini 3.7 Flash 37.8%
  14. 14 Claude Opus 4.8 37.5%
  15. 15 GPT-5.6 Sol 37.2%
  16. 16 Claude Sonnet 5 36.6%
  17. 17 Gemini 3.8 Flash 35.1%
  18. 18 DeepSeek V4.1 Flash 34.8%
  19. 19 GPT-5.5 33.9%
  20. 20 GPT-6 Luna 33.6%
  21. 21 Claude Opus 4.7 31.9%
  22. 22 GLM-5.3 31.9%
  23. 23 Kimi K3 31.3%
  24. 24 GPT-6.1 Terra 30.4%
  25. 25 Claude Sonnet 4.6 30.1%
  26. 26 Beam 1.0 27.4%
  27. 27 Mistral Large 4 27.1%
  28. 28 Nemotron 3 Ultra 25.1%
  29. 29 GPT-5.6 Luna 23%
  30. 30 Nemotron 3.5 Lightning 30B-A3B 8.55%

Source: Official TxBench-Oligo-Discovery leaderboard

#ModelScoreSettingDateSource
1 Claude Opus 5.5 62.2% LSVP, Claude Code Sep 2026 Official leaderboard
2 Gemini 4 Carbon 55.8% Pi Oct 2026 Official leaderboard
3 Claude Fable 5.1 54.9% LSVP, Pi Sep 2026 Official leaderboard
4 Gemini 4 Hydrogen 54.3% Pi Nov 2026 Official leaderboard
5 GPT-6.1 Sol 52.5% Pi Sep 2026 Official leaderboard
6 GPT-6 Astra 51.9% Pi Sep 2026 Official leaderboard
7 Claude Sonnet 5.5 51.9% Claude Code Sep 2026 Official leaderboard
8 Gemini 4 Argon 51% Pi Sep 2026 Official leaderboard
9 Claude Opus 5 48.4% Claude Code Jul 2026 Official leaderboard
10 Grok 4.6 46% Pi Aug 2026 Official leaderboard
11 GPT-6 Sol 43.1% Pi Sep 2026 Official leaderboard
12 Grok 4.7 41.9% Pi Sep 2026 Official leaderboard
13 Gemini 3.7 Flash 37.8% Pi Aug 2026 Official leaderboard
14 Claude Opus 4.8 37.5% Pi May 2026 Official leaderboard
15 GPT-5.6 Sol 37.2% Codex Jul 2026 Official leaderboard
16 Claude Sonnet 5 36.6% Pi Jun 2026 Official leaderboard
17 Gemini 3.8 Flash 35.1% Pi Sep 2026 Official leaderboard
18 DeepSeek V4.1 Flash 34.8% Pi Sep 2026 Official leaderboard
19 GPT-5.5 33.9% Pi Apr 2026 Official leaderboard
20 GPT-6 Luna 33.6% Pi Sep 2026 Official leaderboard
21 Claude Opus 4.7 31.9% Pi Apr 2026 Official leaderboard
22 GLM-5.3 31.9% Pi Aug 2026 Official leaderboard
23 Kimi K3 31.3% Pi Jul 2026 Official leaderboard
24 GPT-6.1 Terra 30.4% Pi Oct 2026 Official leaderboard
25 Claude Sonnet 4.6 30.1% Pi Feb 2026 Official leaderboard
26 Beam 1.0 27.4% Pi Oct 2026 Official leaderboard
27 Mistral Large 4 27.1% Pi Oct 2026 Official leaderboard
28 Nemotron 3 Ultra 25.1% Pi Jun 2026 Official leaderboard
29 GPT-5.6 Luna 23% Codex Jul 2026 Official leaderboard
30 Nemotron 3.5 Lightning 30B-A3B 8.55% Pi Aug 2026 Official leaderboard

2 models have held number one on the official leaderboard since its release in September 2026

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

Anthropic
Own leaderboard 2 leaders
Claude Opus 5.5 62.2%

Source: Official TxBench-Oligo-Discovery leaderboard

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 Claude Fable 5.1 Claude Opus 5 Claude Opus 4.8 Claude Opus 4.7 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Claude Opus 5.5 Claude Fable 5.1

Source: Official TxBench-Oligo-Discovery leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Biology and chemistry

All 109 in Biology and chemistry
  • 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-5.6 Sol 31.5%
  • 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%
  • 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-Rosalind-6 61.3%
  • MedChemBench (OpenAI) In use Models answer chemistry questions from an OpenAI internal set, grouped with OpenAI's life science research evaluations. OpenAI has not published the task format. GPT-Rosalind-6 49.6%
  • ProtocolQA Open-Ended In use Models correct errors in published lab protocols using open-ended questions adapted from LAB-Bench ProtocolQA. Scored as pass@1. GPT-6 Cyber 47.2%
  • MolecularIQ In use Models answer questions about molecular graphs, such as counting, indexing and constrained generation, and a symbolic checker verifies each answer. gpt-oss-120b 47.5%