Reasoning Datasets - a ermiaazarkhalili Collection

ermiaazarkhalili 's Collections

Instruction Datasets

Multimodal Datasets

Reasoning Datasets

Qwen-Function-Calling-xLAM

Llama-Function_Calling-xLAM

VLMs

Mistral-GRPO-GSM8K

Llama-GRPO-GSM8K

Qwen2.5-GRPO-GSM8K

Reasoning Datasets

updated Dec 29, 2025

zwhe99/DeepMath-103K

Viewer • Updated May 29, 2025 • 103k • 9.1k • 351
nvidia/Nemotron-CC-Math-v1

Viewer • Updated Dec 23, 2025 • 190M • 3.69k • 66
AI-MO/NuminaMath-TIR

Viewer • Updated Nov 25, 2024 • 72.5k • 2.77k • 142
openai/gsm8k

Benchmark • Updated Dec 20, 2025 • 17.6k • 471k • 1.16k
microsoft/orca-math-word-problems-200k

Viewer • Updated Mar 4, 2024 • 200k • 4.18k • 474

Note Orca-Math - GPT-4 quality, ~350 tokens avg, 200K samples
meta-math/MetaMathQA

Viewer • Updated Dec 21, 2023 • 395k • 16k • 442

Note MetaMathQA - Augmented GSM8K/MATH, ~500 tokens avg, 395K samples
nvidia/OpenMathInstruct-2

Viewer • Updated Nov 25, 2024 • 22M • 15.5k • 230

Note OpenMathInstruct-2 - NVIDIA, 14M samples, high quality
open-r1/OpenR1-Math-220k

Viewer • Updated Feb 18, 2025 • 450k • 11k • 711

Note OpenR1-Math - R1-style reasoning, 220K samples
AI-MO/NuminaMath-CoT

Viewer • Updated Nov 25, 2024 • 860k • 11.2k • 536

Note NuminaMath-CoT - Competition math, 860K samples