Grok 4.5

Jul 2026
Grok 4.5
Compare in
on
$0.88per task?
$2 / $6per 1M in / outcache $0.30
63 tok/sfirst token in 12 s?
500Kcontext

Rankingat high effort?

Full ranking →
1GPT-6 Astra952GPT-6.1 Sol94.23Opus 5.593.34GPT-5.6 Sol92.55Gemini 4 Argon91.3
···
32GPT-5.252.933Grok 4.552.634Qwen 27B42.4

Strong at

S
Reasoning#33 of 334Science#12 of 312Knowledge#20 of 318
A
Agents & work tasks#40 of 175Long context#53 of 273Vision#19 of 124
B
Coding#39 of 132

Alternatives?

ModelScoreCostPer taskWhat changes
CheaperGPT-6.1 Solmax+79%-18%$0.72Remove unsupported parameters+5BetterGPT-6 Lunamax+13%-92%$0.068Function calling only with none · Chat Completions+3OpenGLM 5.3 Flashmax+25%-68%$0.28Defaults to max · Self-hosted+4CheaperGPT-5.6 Solhigh+62%-8%$0.81CheaperGPT-6 Astralow+62%-7%$0.82Monitoring can return HTTP 403+2

History

8 versions
0204060Jul 25Jan 26Jul 26Release dateARC-AGI-2 · Semi-private344.20 · Apr4.54.64.7
0204060Jul 25Jan 26Jul 26Release dateARC-AGI-2 · Semi-private344.20 · Apr4.54.64.7

Benchmarks

in Chat & Knowledge
AA Intelligence Index 4.3.238.8#35
AA Omniscience Index25.3#20
AA Omniscience · Accuracy51.6#17
AA Omniscience · Hallucination Rate54.1#98
Humanity's Last Exam42.7#33
GPQA Diamond93.1#12
AA-LCR (long context) 1.179.3#53
CritPt (physics)15.4#43
GDP.pdf · AA all-pass18.8#27
MLCR-AA15#34
ARC-AGI-1 · public evaluation94.8#26
ARC-AGI-1 · semi-private85.7#33
ARC-AGI-2 · public evaluation58.2#29
ARC-AGI-2 · semi-private52.6#33
ARC-AGI-3 · semi-private0.3#15
GPQA Diamond · Epoch93.4#13
OTIS Mock AIME 2024 202597.8#26
FrontierMath Tiers 1–3 (v2)57.2#38
FrontierMath Tier 4 (v2)24.4#37
SimpleQA Verified v148.3#29
Chess Puzzles35#38

FAQ

How much does Grok 4.5 cost?›

$2 per 1M input tokens and $6 per 1M output tokens; cached input $0.30 per 1M. A typical benchmark task costs about $0.88 at high effort.

Is Grok 4.5 free to use?›

There is no free API; the cheapest paid price is $2 per 1M input tokens.

How good is Grok 4.5, and what is it best at?›

#33 of 334 in Chat & Knowledge, #39 of 132 in Coding, #24 of 133 in Agents & Tool Use, #24 of 195 in Writing & Language, #19 of 124 in Vision & Documents. Strongest at: Agents & work tasks, Reasoning, Science, Long context, Knowledge, Vision.

Is Grok 4.5 better than GPT-5.2?›

On ARC-AGI-2 · semi-private, Grok 4.5 scores 52.6 vs 52.9 — about the same.

How can I use Grok 4.5?›

Through the API of xAI.

What is the context window of Grok 4.5?›

500K tokens, with up to 450K output tokens.

When was Grok 4.5 released? Is there a newer version?›

Released 2026-07-08. It has been superseded by Grok 4.6.

Is Grok 4.5 open source?›

No, Grok 4.5 is a closed model available through an API. The closest open-weights option here is GLM 5.3 Flash.

Can Grok 4.5 read images?›

It accepts text and images, PDF files as input.

What is Grok 4.5?›

Grok 4.5 is a closed language model from xAI, released Jul 2026, the successor to Grok Build 0.1.

What are the best alternatives to Grok 4.5?›

GPT-6.1 Sol (cheaper), GPT-6 Luna (better), GLM 5.3 Flash (open weights).