Anthropic Protein Design evaluation (Protein Design)
We show this table for reference; we do not rank on it.
Generates novel protein sequences under family, topology, globularity, and structural-motif constraints.
Benchmark score on Protein Design — November 11, 2026
We compile the Protein Design rows from provider self-reports. Claude Mythos 5.2 leads the table at 64.6%, followed by Claude Opus 5.5 (60.2%) and Claude Sonnet 5.5 (51.0%). We do not use these results to rank models overall.
Claude Mythos 5.2
Anthropic
Claude Opus 5.5
Anthropic
Claude Sonnet 5.5
Anthropic
4 modelsKnowledgeCurrentDisplay onlyUpdated November 11, 2026
Benchmark score table (4 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Claude Mythos 5.2Anthropic | 64.6% | Not reported | Closed |
| 2 | Claude Opus 5.5Anthropic | 60.2% | Not reported | Closed |
| 3 | Claude Sonnet 5.5Anthropic | 51.0% | Not reported | Closed |
| 4 | Claude Opus 5Anthropic | 42.5% | Not reported | Closed |
Among the reported Protein Design rows, Claude Mythos 5.2 is first at 64.6%. The third row is 13.6 points behind. The broader top-10 range is 22.1 points, so the table still separates the published systems.
4 models have been evaluated on Protein Design. The benchmark falls in the Knowledge category. Protein Design is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About Protein Design
Year
2026
Tasks
Constrained protein-sequence design tasks
Format
Composite score
Difficulty
Computational protein design
Section 8.17.4 reports a composite of constraint satisfaction, folding confidence, and sequence novelty without tool access.
Freshness and provenance
Version
Protein Design 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does Protein Design measure?
Generates novel protein sequences under family, topology, globularity, and structural-motif constraints.
Which model scores highest on Protein Design?
Claude Mythos 5.2 by Anthropic currently leads with a score of 64.6% on Protein Design.
How many models are evaluated on Protein Design?
4 AI models have published results on Protein Design in the BenchLM catalog.
Compare top models on Protein Design
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.