Grok 3

Active

xAI's frontier reasoning model with real-time web access.

Overview

Grok 3 is xAI's most capable model with strong reasoning capabilities and real-time access to information through X's data firehose. Available through the xAI API and to X Premium subscribers.

Benchmarks

BenchmarkScoreSource
AIME 2024Math83.9% accuracySelf-reported
xAI model card
Aider PolyglotCoding53.3% pass@2Third-party
Papers With Code
GPQA DiamondReasoning84.6% accuracySelf-reported
xAI model card
HumanEvalCoding88.5pass@1 %Self-reported
xAI model card
MATHMath93.3% accuracySelf-reported
xAI model card
MMLUGeneral knowledge87% accuracySelf-reported
xAI model card
MMLU-ProGeneral knowledge79.3% accuracySelf-reported
xAI model card
SWE-bench VerifiedCoding50% resolvedThird-party
Papers With Code

Integrations & tooling support

Tool calling
Supported
Structured outputs
Supported

Price vs quality

Overpriced

Mid-tier performance at frontier pricing.

Quality percentile
58.9%
vs 8 benchmarks
Effective price
$12/1M
/ 1M tokens (input + 3× output)
Pricing breakdown
$3/1M in
$15/1M out

Community ratings

No ratings yet. Be the first to rate Grok 3.

Rate Grok 3

Sign in to rate and review.

Comments

Sign in to leave a comment.