#18xAI· released 2026-04-30

xAI: Grok 4.3.

Grok 4.3 is a reasoning model from xAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual...

View on OpenRouter ↗
Context
1.0M
Input $/1M
$1.25
Output $/1M
$2.50
Modality
text · image · file

Benchmarks

How these are aggregated →
Arena Elo
1473
LMArena — human blind votes
AA Index
58.0
Artificial Analysis composite
LiveBench
73.9
Contamination-resistant
GPQA
81.7%
Grad-level science reasoning
MMLU-Pro
83.6%
Broad knowledge
SWE-bench
65.8%
Real GitHub bugs (Verified)
Terminal-Bench
45.2%
Agentic shell tasks
ARC-AGI
14.9%
Abstract reasoning
Aider Polyglot
76.4%
Multi-language code editing
Speed
109 tok/s
Median output speed

Nearby on the leaderboard