Cannae-AI
/

Qwen3-MATH-R1-4B

Text Generation

text-generation-inference

Model card Files Files and versions

Qwen3-MATH-R1-4B

Model Description

This is a fine-tuned version of Qwen/Qwen3-4B-Thinking-2507 on parts of the nvidia/OpenMathReasoning dataset which was used to win the AIMO (AI Mathematical Olympiad) challenge!

recommended settings for instruct inference: temperature = 0.7, top_p = 0.8, top_k = 20
For reasoning chat based inference : temperature = 0.6, top_p = 0.95, top_k = 20
License : apache-2.0
Finetuned from model : Qwen/Qwen3-4B-Thinking-2507

Downloads last month: 3

Safetensors

Model size

4B params

Tensor type

BF16

·

Model tree for Cannae-AI/Qwen3-MATH-R1-4B

Base model

Qwen/Qwen3-4B-Thinking-2507

Finetuned

(115)

this model

Quantizations

1 model

Dataset used to train Cannae-AI/Qwen3-MATH-R1-4B

Collection including Cannae-AI/Qwen3-MATH-R1-4B

Math + Coding LMs

A collection of fine tuned models spanning mathematics, coding, and cybersecurity.Engineered for comprehensive coverage of computational reasoning. • 6 items • Updated 2 days ago • 1