Grok 4.6

ModelActive

by xAI · family “grok”

Grok 4.6 is a model from SpaceXAI with frontier performance on coding, knowledge work, and STEM. It is succeeded by [Grok 4.7](/x-ai/grok-4.7). (description from the OpenRouter listing)

in: textin: imagein: fileout: textReasoningTool useStructured output

Epoch Capabilities Index

156.5 #18 of 268

90% CI 154.6 – 159.0

Listing ↗

Released

12 Aug 2026

Context window

500K tokens

Max output

450K tokens

Input $ / 1M tokens

$2.00

Output $ / 1M tokens

$6.00

Cached input $ / 1M

$0.50

Each benchmark shown separately with its own source

Benchmark results (8)

All benchmarks

Benchmark	Domain	Score	vs best recorded	Setting	Run	Source
DeepSWE	coding	65.2%	88%	high	—	External ↗
Chess Puzzles	games	40.0% ±4.9	56%	high	12 Aug 2026	Epoch ↗
SimpleQA Verified	knowledge	49.3% ±1.6	65%	high	27 Aug 2026	Eval log ↗
ARC-AGI-2	reasoning	65.1%	69%	high	—	External ↗
WeirdML	coding	67.3% ±0.0	72%	high	—	External ↗
OTIS Mock AIME 2024-2025	math	97.8% ±1.0	98%	high	12 Aug 2026	Epoch ↗
GPQA diamond	science	94.0% ±1.4	98%	high	12 Aug 2026	Epoch ↗
ARC-AGI	reasoning	87.0%	88%	high	—	External ↗

Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.

API list price over time

Price history

$ per 1M tokens

2 observations (OpenRouter listing + Internet Archive snapshots). History accumulates with every ingest run; a single point means no change has been observed yet.