SATBench
Fresh
SATBench is a benchmark for evaluating the logical reasoning capabilities of large language models (LLMs) through logical puzzles derived from Boolean satisfiability (SAT) problems.
- Type
- RL Env
- Runtime
ORS- License
- unknown
- Size
- 2100 tasks
- Published
- Mar 2026
- Canonical
- openreward.ai/anjiang/SATBench
Cite
Notes
Only stored in your browser.