---
title: "Grok 4.3 Benchmarks - RankedAGI"
description: "Detailed benchmark and metadata record for Grok 4.3."
source: "https://rankedagi.com/models/grok-4-3"
---

# Grok 4.3

| Field | Value |
| --- | --- |
| Organization | xAI |
| License | Proprietary |
| Type | Thinking |
| Version | Latest |
| Released | 2026-04-30 |
| Context window | 1M |
| Input cost per million tokens | $1.25 |
| Output cost per million tokens | $2.5 |
| Cached input cost per million tokens | $0.2 |
| Last updated | 2026-05-04T14:41:26.631Z |

## Benchmarks

| Benchmark | Category | Value | Description | Source |
| --- | --- | --- | --- | --- |
| RankedAGI Coding | coding | 72.8% | RankedAGI Coding Score |  |
| RankedAGI Agentic | Agents | 62.5% | RankedAGI Agentic Score |  |
| Code Arena | coding | 1408 | Code Arena Elo Score | https://arena.ai/leaderboard/code |
| SvelteBench | coding | 94.4% | SvelteBench - Benchmark for Svelte | https://khromov.github.io/svelte-bench/benchmark-results-merged.html |
| RankedAGI Reasoning | reasoning | 71.9% | RankedAGI Reasoning Score |  |
| Humanity's Last Exam | reasoning | 35.0% | Multidisciplinary Reasoning (no tools) |  |
| GPQA Diamond | reasoning | 90.0% | Generalized Prefix Question Answering Score (Reasoning) PhD Level Reasoning |  |
| Text Arena | general | 1456 | ChatArena (LMSYS) ELO Score | https://arena.ai/leaderboard/text |
| Vending Bench 2 | Agents | $704.98 | Benchmark for measuring AI model performance on running a business over long time horizons. Models are tasked with running a simulated vending machine business over a year and scored on their bank account balance at the end. | https://andonlabs.com/evals/vending-bench-2 |
| Instruction Following Evaluation | instruction-following | 81.0% | Instruction Following Evaluation |  |
| MMMU-Pro | imaging | 78.0% | Multimodal Understanding |  |
| RankedAGI Math | math | 71.9% | RankedAGI Math Score |  |
| RankedAGI Overall | general | 63.8% | Overall RankedAGI score |  |
| 𝜏²-Bench Telecom | Agents | 98.0% | Agentic Tool Use |  |
| GDPval-AA Elo | Agents | 1500 | Office Tasks (Artificial Analysis) |  |