ModelBeat
Models/Grok 4.5
All models
X
x-ai/grok-4.5

Grok 4.5 is xAI’s proprietary AI model, released Jul 8, 2026. It scores 155 on the Epoch Capabilities Index, costs $2/M input and $6/M output tokens, and handles a 500K-token context window. Its strongest use case is coding, where it ranks in the 73rd percentile of the models Model Beat tracks.

Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM.

Context
500K
Input
$2 / 1M
Output
$6 / 1M
Released
Jul 8, 2026
Modalities
textimagefile→text
Strongest atCoding top 27%Agentic & Tools top 30%Math top 39%

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Grok 4.5 lands among all models tracked on Model Beat.

64.0
Intelligence Index
Epoch AI
64th percentile of tracked models
73.0
Coding Index
Epoch AI
73rd percentile of tracked models
70.0
Agentic Index
Epoch AI
70th percentile of tracked models

Reasoning

7 evals
ARC-AGI87.2%

Abstract visual reasoning

ARC-AGI-252.6%

Harder abstract reasoning

GPQA Diamond93.4%

Graduate-level scientific reasoning

Humanity's Last Exam42.7%

Frontier of human expert knowledge

SimpleBench70.0%

Common-sense trick questions

SimpleQA Verified48.3%

Factual accuracy & hallucination

WeirdML46.4%

Novel ML problem-solving

Coding

2 evals
SciCode55.0%

Scientific research coding

WebDev Arena1555

Human-rated web development

Math

1 evals
AIME 2024/202597.8%

Olympiad-qualifier math

Agentic & Tools

1 evals
APEX56.2%

Multi-step agentic tasks

In the news

2

Changelog

4
  • Grok 4.5: reported APEX score rose from 34.2% to 56.2% (+64%)benchmark · Sep 22, 2026
  • Grok 4.5: reported SimpleQA Verified score fell from 53.5% to 48.3% (-10%)benchmark · Aug 28, 2026
  • Grok 4.5: Humanity's Last Exam improved from 40.3% to 42.7% (+6%)benchmark · Aug 6, 2026
  • Grok 4.5 (xAI) added to the tracker.tracked · Jul 8, 2026

Compare Grok 4.5 with

Frequently asked questions

What is Grok 4.5?

Grok 4.5 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM.

Who created Grok 4.5?

Grok 4.5 is an AI model developed by xAI, part of the Grok family, released Jul 8, 2026.

How much does Grok 4.5 cost?

Grok 4.5's list price from xAI is $2 per million input tokens and $6 per million output tokens.

What is Grok 4.5's context window?

Grok 4.5 supports a context window of up to 500K tokens.

How does Grok 4.5 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Grok 4.5 scores 93.4% on GPQA Diamond, 55.0% on SciCode, 42.7% on Humanity's Last Exam.

Is Grok 4.5 open source?

Grok 4.5 is a proprietary model (API access), available through its provider's API rather than as a downloadable open-weight model.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Nov 19, 2026.