Black-box RNA sequence modeling and design
Models predict scores for unlabeled RNA sequences from labeled data, then design new sequences. Results are compared with 57 human experts.
Built with Dyno Therapeutics for the CB-2 threshold. Reported as design and rank-correlation prediction scores in charts.
- Released
- May 2026
- Kind
- Hazardous knowledge
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Kept private
- Last in a system card
- October 2026
- Last model added
- October 2026
8
System cards report it, from 1 lab
No scores here yet
No lab, leaderboard or Epoch AI run has reported a score for Black-box RNA sequence modeling and design yet.
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
8 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Oct 7, 2026 | Claude Haiku 5.5 system card | Black-box RNA sequence modeling and design (Dyno), design and prediction scores, chart without printed values (Figures 2.2.3.1.A and 2.2.3.1.B) | Bio and chem risk | |
| Sep 28, 2026 | Claude Sonnet 5.5 system card | Black-box RNA sequence modeling and design, design and prediction scores, chart without printed values (2.2.3.1, Figures 2.2.3.1.A and B) | Bio and chem risk | |
| Sep 22, 2026 | Claude Opus 5.5 system card | Black-box RNA sequence modeling and design (Dyno), design and prediction scores, chart without printed values. Text says the top design score exceeds the 75th human percentile (2.2.3.2.1, Figures 2.2.3.2.1.A and B) | Bio and chem risk | |
| Jul 24, 2026 | Claude Opus 5 system card | Black-box RNA sequence modeling and design (Dyno), top and median design scores and prediction scores, baseline and in-context iteration, chart without printed values. Text says it exceeds the 75th human percentile on design and prediction (2.2.5.1, Figures 2.2.5.1.A and B) | Bio and chem risk | |
| Jun 30, 2026 | Claude Sonnet 5 system card | Black-box RNA sequence modeling and design (Dyno), design and prediction scores over eight attempts, chart without printed values. Text: prediction score above the 90th human percentile (2.2.5.1, Figures 2.2.5.1.A and B) | Bio and chem risk | |
| May 28, 2026 | Claude Opus 4.8 system card | Black-box RNA sequence modeling and design (Dyno), eight attempts, no overall: design score top 10.06 and median 6.84, prediction score (Spearman) all sequences 0.786 and top 5% -0.029, no prior context (Table 2.2.3.B, 2.2.5.1) | Bio and chem risk | |
| Apr 16, 2026 | Claude Opus 4.7 system card | Sequence-to-function modeling and design (Dyno), prediction and design scores, chart without printed values. Text says prediction exceeds the 90th human percentile and design stays below the 75th (2.2.5.3, Figure 2.2.5.3.A) | Bio and chem risk | |
| Apr 7, 2026 | Claude Mythos Preview system card | Sequence-to-function modeling and design (Dyno Therapeutics), prediction and design scores over 8 attempts, chart without printed values. Text says the mean beats the 75th human percentile on both and the 90th on prediction, below the top human (2.2.5.5, Figure 2.2.5.5.A) | Bio and chem risk |
More 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%