Grok 4.6: benchmarks, price and rank
Grok 4.6 from xAI is 26th of 123 rated AI models on Spezia's leaderboard. It costs $2 in, $6 out per million tokens.
- 26th of 123 rated models
- Rating 1,532
- Rated on 13 benchmarks
- Closed
- Released Aug 12, 2026
- 500,000 tokens of context
- Knows the world to February 2026
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 16th of 21 on FrontierSWE
- 15th of 20 on CursorBench
- 18th of 27 on DeepSWE 1.1
Price: $2 in, $6 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- APEX-Agents65.3%
Long tasks written by investment banking analysts, management consultants and corporate lawyers from their own work, done across files and apps in a realistic work setup.
13th of 56. Independent result from Epoch AI.
- Terminal-Bench 4.020.3%
Hard jobs done in a real computer terminal: building and fixing software, setting up systems, and science and engineering problems.
Independent result from LLM Stats.
- FrontierSWE25.3%
Open-ended technical projects that take hours to tens of hours, from making systems faster to building large programs and applied machine learning research. Each run gets a 20-hour budget.
16th of 21. Independent result from Epoch AI.
- Vending-Bench 2$9,047
The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.
14th of 60. Independent result from Epoch AI.
- CursorBench41.4%
Ambiguous coding tasks that span many files, taken from real sessions in the Cursor code editor.
15th of 20. Independent result from Epoch AI.
- DeepSWE 1.165.9%
Software engineering tasks run in the same simple agent setup for every model.
18th of 27. The maker's own number, from xAI, via LLM Stats.
- SimpleBench75.9%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
16th of 68. Independent result from Epoch AI.
- SimpleQA Verified49.3%
Short factual questions with one checked answer, asked without web search.
32nd of 78. Independent result from Epoch AI.
- FrontierMath66.0%
Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.
36th of 76. Independent result from Epoch AI.
- LMArena text1,430
People put a question to two anonymous models and vote for the better answer.
63rd of 107. Independent result from LMArena.
6 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.