Benchmark Almanac

TxBench-Oligo-Discovery

Claude Opus 5.5 62.2% leads since Sep 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 Sep 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-oligo-discovery 0 stars Leaderboard benchmarks.bio
28
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 2 Nov 2026. Where a model ran with several settings or harnesses, we keep its best.

28 models. Scroll sideways for the rest.

  1. 62.2 Claude Opus 5.5
  2. 54.9 Claude Fable 5.1
  3. 52.5 GPT-6.1 Sol
  4. 51.9 GPT-6 Astra
  5. 51.9 Claude Sonnet 5.5
  6. 51.0 Gemini 4 Argon
  7. 48.4 Claude Opus 5
  8. 46.0 Grok 4.6
  9. 43.1 GPT-6 Sol
  10. 41.9 Grok 4.7
  11. 37.8 Gemini 3.7 Flash
  12. 37.5 Claude Opus 4.8
  13. 37.2 GPT-5.6 Sol
  14. 36.6 Claude Sonnet 5
  15. 35.1 Gemini 3.8 Flash
  16. 34.8 DeepSeek V4.1 Flash
  17. 33.9 GPT-5.5
  18. 33.6 GPT-6 Luna
  19. 31.9 Claude Opus 4.7
  20. 31.9 GLM-5.3
  21. 31.3 Kimi K3
  22. 30.4 GPT-6.1 Terra
  23. 30.1 Claude Sonnet 4.6
  24. 27.4 Beam 1.0
  25. 27.1 Mistral Large 4
  26. 25.1 Nemotron 3 Ultra
  27. 23.0 GPT-5.6 Luna
  28. 8.55 Nemotron 3.5 Lightning 30B-A3B
  1. 1 Claude Opus 5.5 LSVP 62.2%
  2. 2 Claude Fable 5.1 LSVP 54.9%
  3. 3 GPT-6.1 Sol 52.5%
  4. 4 GPT-6 Astra 51.9%
  5. 5 Claude Sonnet 5.5 51.9%
  6. 6 Gemini 4 Argon 51%
  7. 7 Claude Opus 5 48.4%
  8. 8 Grok 4.6 46%
  9. 9 GPT-6 Sol 43.1%
  10. 10 Grok 4.7 41.9%
  11. 11 Gemini 3.7 Flash 37.8%
  12. 12 Claude Opus 4.8 37.5%
  13. 13 GPT-5.6 Sol 37.2%
  14. 14 Claude Sonnet 5 36.6%
  15. 15 Gemini 3.8 Flash 35.1%
  16. 16 DeepSeek V4.1 Flash 34.8%
  17. 17 GPT-5.5 33.9%
  18. 18 GPT-6 Luna 33.6%
  19. 19 Claude Opus 4.7 31.9%
  20. 20 GLM-5.3 31.9%
  21. 21 Kimi K3 31.3%
  22. 22 GPT-6.1 Terra 30.4%
  23. 23 Claude Sonnet 4.6 30.1%
  24. 24 Beam 1.0 27.4%
  25. 25 Mistral Large 4 27.1%
  26. 26 Nemotron 3 Ultra 25.1%
  27. 27 GPT-5.6 Luna 23%
  28. 28 Nemotron 3.5 Lightning 30B-A3B 8.55%
#ModelScoreSettingDateSource
1 Claude Opus 5.5 62.2% LSVP, Claude Code Sep 2026 Official leaderboard
2 Claude Fable 5.1 54.9% LSVP, Pi Sep 2026 Official leaderboard
3 GPT-6.1 Sol 52.5% Pi Sep 2026 Official leaderboard
4 GPT-6 Astra 51.9% Pi Sep 2026 Official leaderboard
5 Claude Sonnet 5.5 51.9% Claude Code Sep 2026 Official leaderboard
6 Gemini 4 Argon 51% Pi Sep 2026 Official leaderboard
7 Claude Opus 5 48.4% Claude Code Jul 2026 Official leaderboard
8 Grok 4.6 46% Pi Aug 2026 Official leaderboard
9 GPT-6 Sol 43.1% Pi Sep 2026 Official leaderboard
10 Grok 4.7 41.9% Pi Sep 2026 Official leaderboard
11 Gemini 3.7 Flash 37.8% Pi Aug 2026 Official leaderboard
12 Claude Opus 4.8 37.5% Pi May 2026 Official leaderboard
13 GPT-5.6 Sol 37.2% Codex Jul 2026 Official leaderboard
14 Claude Sonnet 5 36.6% Pi Jun 2026 Official leaderboard
15 Gemini 3.8 Flash 35.1% Pi Sep 2026 Official leaderboard
16 DeepSeek V4.1 Flash 34.8% Pi Sep 2026 Official leaderboard
17 GPT-5.5 33.9% Pi Apr 2026 Official leaderboard
18 GPT-6 Luna 33.6% Pi Sep 2026 Official leaderboard
19 Claude Opus 4.7 31.9% Pi Apr 2026 Official leaderboard
20 GLM-5.3 31.9% Pi Aug 2026 Official leaderboard
21 Kimi K3 31.3% Pi Jul 2026 Official leaderboard
22 GPT-6.1 Terra 30.4% Pi Oct 2026 Official leaderboard
23 Claude Sonnet 4.6 30.1% Pi Feb 2026 Official leaderboard
24 Beam 1.0 27.4% Pi Oct 2026 Official leaderboard
25 Mistral Large 4 27.1% Pi Oct 2026 Official leaderboard
26 Nemotron 3 Ultra 25.1% Pi Jun 2026 Official leaderboard
27 GPT-5.6 Luna 23% Codex Jul 2026 Official leaderboard
28 Nemotron 3.5 Lightning 30B-A3B 8.55% Pi Aug 2026 Official leaderboard

2 models have held number one since its release in Sep 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%

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

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%