LitBench
Literary evaluation benchmark using pair-wise comparison with self-critique prompting
- Domain
- rl-env
- License
- unknown
- Published
- Oct 2025
Cite
Notes
Only stored in your browser.
Top score 80.0% by GLM 4.5 Air - 7 models reporting (2 frontier)
Score history
4Top models
7
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is LitBench?
- Literary evaluation benchmark using pair-wise comparison with self-critique prompting
- What is the current top score on LitBench?
- The top reported score is 80.0% by GLM 4.5 Air, across 7 models reporting (2 from frontier labs).
- How can a model improve its LitBench score?
- Tools linked to LitBench on Sophon include Litbench RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
- What license is LitBench under?
- LitBench is available under unknown.