Artificial Analysis
#9of 111
Score 38.8. Best run: x-ai/grok-4.5.
Intelligence Index: a composite of reasoning, knowledge, maths and coding evaluations.
xAI
Grok 4.5 benchmarks, ranks and price
Where Artificial Analysis, LiveBench, LMArena and Berkeley BFCL place Grok 4.5, what it costs through OpenRouter and Vercel AI Gateway, and the models either side of it on the consensus.
Read Oct 12, 2026
#12of 122
on the LLM leaderboard, with a consensus score of 75 across 3 leaderboards.
Source: Artificial Analysis, LiveBench, LMArena and Berkeley BFCL, combined by Rank.ai.
By leaderboard
Where each leaderboard places Grok 4.5
LiveBench
#10of 36
Score 76.2. Best run: Grok 4.5.
Contamination-limited questions refreshed each release: reasoning, maths, coding, data analysis, language.
LMArena
#21of 112
Score 1468. Best run: grok-4.5.
Blind head-to-head human votes, style controlled.
Berkeley BFCL
Not measured by Berkeley BFCL.
Function calling and tool use, single turn to multi turn and web search.
Price and context
$3.00
OpenRouter: $2.00 in, $6.00 out Context window 500K tokens. LiveBench cost per solved task $0.128.
Nearby
The models either side of Grok 4.5
- #9GPT-5.479
- #10Gemini 3.1 Pro Preview77
- #11Muse Spark 1.176
- #12Grok 4.575
- #13Claude Opus 4.673
- #14Claude Sonnet 573
- #15Gemini 3.5 Flash72
Questions about Grok 4.5
How good is Grok 4.5?
Grok 4.5 ranks #12 of 122 models on the Rank.ai LLM leaderboard, with a consensus score of 75 out of 100 across 3 leaderboards.
Is Grok 4.5 better than Muse Spark 1.1?
Not on the consensus: Muse Spark 1.1 scores 76 to Grok 4.5's 75. Compare them source by source below, since the leaderboards test different things.
Is Grok 4.5 better than Claude Opus 4.6?
On the consensus, yes: 75 to 73.
How much does Grok 4.5 cost?
$2.00 per million input tokens and $6.00 per million output tokens on OpenRouter, the cheaper listing we read. Blended at three input tokens to one output, that is $3.00 per million.
Is Grok 4.5 good for coding?
It scores 72.4 on the Artificial Analysis Coding Index, placing it #9 for coding among the models on this leaderboard.
These models recommend brands. See if they name yours.
Enter your website. In about two minutes, Rank.ai asks ChatGPT, Claude and Gemini 12 questions your customers ask and grades how often they name you.
- Your grade out of 100How often AI names you, cites your site, and how high it ranks you.
- Who gets namedEvery competitor in the answers, most named first.
- The pages AI readsThe sources behind each answer.
- Three fixesWhat to fix first, with a brief for the first article.