RiemannBench without tools (RiemannBench (no tools))
We show this table for reference; we do not rank on it.
Research-level mathematics problems with unique programmatically verified closed-form answers.
Benchmark score on RiemannBench (no tools) — September 27, 2026
We compile the RiemannBench (no tools) rows from provider self-reports. Claude Opus 5 leads the table at 60.0%. We do not use these results to rank models overall.
1 modelMathematicsCurrentDisplay onlyUpdated September 27, 2026
Benchmark score table (1 model)
ScoreAbout RiemannBench (no tools)
Year
2026
Tasks
25 private research-level mathematics problems
Format
Accuracy without tools
Difficulty
Research mathematics
Section 8.7 reports the mean over four max-effort attempts on a corrected private 25-problem setup without tools.
Freshness and provenance
Version
RiemannBench (no tools) 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 RiemannBench (no tools) measure?
Research-level mathematics problems with unique programmatically verified closed-form answers.
Which model scores highest on RiemannBench (no tools)?
Claude Opus 5 by Anthropic currently leads with a score of 60.0% on RiemannBench (no tools).
How many models are evaluated on RiemannBench (no tools)?
1 AI models have been evaluated on RiemannBench (no tools) on BenchLM.
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 2,000+ readers.
One email each week. Unsubscribe anytime.