Grok 4.6 is SpaceXAI's smartest model with frontier performance on coding, knowledge work, and STEM.
Tested on 20 benchmarks · BenchGecko score 66.2. Top scores: OTIS Mock AIME 2024-2025 (99.2%), Dtbench (95.5%), GPQA diamond (92.0%).
DeepSeek V4 Pro 0813 scores 66.8 (101% as good) at $0.55/1M input · 73% cheaper
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.
Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.
ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
- Typemultimodal
- Context500K tokens (~250 books)
- ReleasedAug 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.010
Frequently Asked Questions
Key facts · as of 2026-10-05
- Grok 4.6 by xAI. BenchGecko score 66.2, rank 53 of 312 scored models (normalized average of public benchmark scores).
- List price $2.00 input · $6.00 output per 1M tokens (as of 2026-10-05).
- Sold by 6 providers (as of 2026-10-05): xAI $2.00 in / $6.00 out · xAI $2.00 in / $6.00 out · Amazon Bedrock $2.20 in / $6.60 out · xAI $2.20 in / $6.60 out · xAI $4.00 in / $12.00 out · and 1 more. Every provider
- Gecko Tests: Who Are You B (Knows who made it) · Tokenizer Tax C (50% more tokens outside English).
How to cite · data as of 2026-10-05
Grok 4.6 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/grok-4-6
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP