Grok 4.3 is xAI's reasoning model for text and image inputs, built for agentic workflows, instruction following, factual accuracy, long-document analysis, and deep research. Reasoning is always active and requests above 200k total tokens are charged at the higher long-context rate. Content policy rejections can still be charged: xAI may pass through a $0.05 moderation-failure fee, or a $0.055 usage-guidelines violation fee, depending on which rejection upstream returns.
Added Apr 30, 2026
Context Window
1.0M
Max Output
1.0M
Input Price (Auto)
$1.25/1M
Output Price (Auto)
$2.50/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
31.0
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Coding Index
25.1
Agentic Index
65.9
GPQA Diamond
Graduate-level scientific reasoning
65.8%
Better than 55% of models compared
HLE
Humanity's Last Exam
6.5%
Better than 58% of models compared
IFBench
Instruction-following benchmark
47.6%
Better than 64% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
65.8%
Better than 67% of models compared
AA-LCR
Long context reasoning evaluation
24.7%
Better than 44% of models compared
GDPval-AA
Economically valuable tasks
49.8%
Better than 96% of models compared
CritPt
Research-level physics reasoning
8.0%
Better than 96% of models compared
SciCode
Python programming for scientific computing
37.4%
Better than 74% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
18.9%
Better than 66% of models compared
AA-Omniscience Accuracy
Proportion of correctly answered questions
34.6%
Better than 97% of models compared
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
25.0%
Better than 98% of models compared
Last updated May 15, 2026
Artificial Analysis