0

Deepscaler Math

Single-turn DeepScaleR math problems with boxed numeric answers and CoT.

Domain
rl-env
License
apache-2.0
Published
Aug 2025

Cite

Notes

Only stored in your browser.

Attribution

Leaderboard scores
prime-hub
Attribution policy →

Top score 80.0% by GPT-4.1 Mini - 2 models reporting (2 frontier)

Score history

2
75%81%88%94%100%Apr 25May 25Jun 25Jul 25Aug 25GPT-4.1 Mini

Top models

2
Deepscaler MathBar chart with 2 bars. Highest value: GPT-4.1 Mini at 80.
2 models

Related tools

1
View all

Implementations, trainers, datasets and scaffolds linked to this eval.

FAQ

What is Deepscaler Math?
Single-turn DeepScaleR math problems with boxed numeric answers and CoT.
What is the current top score on Deepscaler Math?
The top reported score is 80.0% by GPT-4.1 Mini, across 2 models reporting (2 from frontier labs).
How can a model improve its Deepscaler Math score?
Tools linked to Deepscaler Math on Sophon include Deepscaler MATH RL Env (Prime Intellect) - RL environments, datasets, and scaffolds that target this eval.
What license is Deepscaler Math under?
Deepscaler Math is available under apache-2.0.