---
title: "Grok 3 mini (Think) Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Grok 3 mini (Think)."
source: "https://rankedagi.com/models/grok-3-mini-think-beta"
---

# Grok 3 mini (Think)

| Field | Value |
| --- | --- |
| Organization | xAI |
| License | Proprietary |
| Version | Preview |
| Released | 2025-02-17 |
| Context window | 1M |
| Knowledge cutoff | 2024-11-17 |
| Last updated | 2026-05-03T23:35:38.868Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 47.7% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 37.0% | RankedAGI Agentic Score |  |
| RankedAGI Reasoning | reasoning | 56.6% | RankedAGI Reasoning Score |  |
| GPQA Diamond | reasoning | 84.0% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1362 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| AIME 2025 | math | 90.8% | AIME 2025 Competition Math |  |
| AIME 2024 | math | 95.8% | AIME 2024 Competition Math |  |
| RankedAGI Math | math | 71.1% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 49.7% | Overall RankedAGI score |  |
| GDPval-AA Elo | Agents | 423 | Office Tasks (Artificial Analysis) |  |

## Model Links

| Title | URL |
| --- | --- |
| Release | https://x.ai/blog/grok-3 |