Compare models

Context, limits, capabilities, price and use cases side by side — up to 4 models at a time. The URL holds your selection, so you can share it.

Grok 4.6 Add a modelChat with this model

Attribute	Grok 4.6x-ai/grok-4.6
Specs
Context window	500K
Max output	500K
Input modalities	TextImage
Output modalities	Text
Capabilities
Streaming	Supported
Tool calling	Supported
Structured output (JSON)	Supported
Image input	Supported
Thinking mode	Supported
Price / 1M tokens
Input	$ 2.00>= 200K$ 4.00
Output	$ 6.00>= 200K$ 12.00
Cache read	$ 0.50>= 200K$ 1.00
Cache write	$ 2.00>= 200K$ 4.00
What it is good at
Best for	Long-running coding agents Knowledge work Interactive and visual builds
Published benchmarks
CursorBench v3.2score	69.9%effort high(setup not fully disclosed)
FrontierCode v1.1 (Extended)score	61.3%effort high(setup not fully disclosed)
Terminal-Bench 3.0pass_rate	26%effort high(setup not fully disclosed)
APEX-Agentsscore	57.5%effort high(setup not fully disclosed)
APEX-SWEscore	56.4%effort high(setup not fully disclosed)
DeepSWE 1.1resolved_rate	65.9%effort high(setup not fully disclosed)
AA-Briefcasescore	1577effort high(setup not fully disclosed)
GDPval-AA v2elo	1753effort high(setup not fully disclosed)
Harvey LAB (Vals)score	15.8%effort high(setup not fully disclosed)
AA Intelligence Indexscore	61effort high(setup not fully disclosed)