Skip to main content
Back to Explore

Grok 3 mini Reasoning (high)

xAI·Released 2025-02-19
MoEMultimodal

Pricing

Input

$0.30

per 1M tokens

Output

$0.50

per 1M tokens

Blended

$0.35

per 1M tokens

Cheaper than 59% of models. Median price is $0.54/1M tokens.

Cost Calculator

Tokens per day1M
100K100M

Daily

$0.35

Monthly

$10.50

vs. Similar Models

Grok 4.20 0309 (Non-reasoning)Q:0.0
$3.00+757%
Gemini 2.5 Pro Preview (May' 25)Q:-0.2
$3.44+882%
Qwen3 235B A22B 2507 (Reasoning)Q:-0.2
$0.84+139%
DeepSeek V3.2 SpecialeQ:-0.3
$0.32-8%

Performance

56

tokens/sec

Faster than 22% of models

0.64

seconds

Faster than 75% of models

36.15

seconds

Faster than 13% of models

Market Median

94 tok/s

40% slower

Median TTFT

1.13s

43% faster

Throughput/Dollar

161

tok/s per $/1M

Speed Comparison

Qwen3 Max (Preview)
57 tok/s+1%
Qwen2.5 Instruct 72B
56 tok/s-1%
GPT-5.5 (Non-reasoning)
57 tok/s+2%

Benchmarks

MMLU-Pro
82.8%
GPQA Diamond
79.1%
HLE
11.1%
LiveCodeBench
69.6%
SciCode
40.6%
TerminalBench Hard
17.4%
MATH-500
99.2%
AIME
93.3%
AIME 2025
84.7%
IFBench
45.9%
Long Context Recall
50.3%
Tau2
90.4%
Market AverageTop Score

Quick Compare

Similar Models

Compare all 7 models