About
Grok Code Fast 1 is a speedy and economical reasoning model that excels at agentic coding. With reasoning traces visible in the response, developers can steer Grok Code for high-quality...
Related Models
Pricing
Input
$0.20
per 1M tokens
Output
$1.50
per 1M tokens
Blended
$0.53
per 1M tokens
Cheaper than 51% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.53
Monthly
$15.75
vs. Similar Models
Grok 4.20 0309 v2 (Non-reasoning)Q:+0.2
$3.00+471%
Nova 2.0 Pro Preview (medium)Q:+0.2
$3.44+555%
DeepSeek: DeepSeek V3.1 TerminusQ:-0.2
$0.44-16%
DeepSeek V3.2 Exp (Non-reasoning)Q:-0.3
$0.31-41%
Performance
79
tokens/sec
Faster than 39% of models
6.25
seconds
Faster than 15% of models
6.25
seconds
Faster than 45% of models
Market Median
94 tok/s
17% slower
Median TTFT
1.11s
462% slower
Throughput/Dollar
150
tok/s per $/1M
Speed Comparison
DeepSeek V3.2 (Reasoning)
79 tok/s+0%
GPT-5 (low)
80 tok/s+1%
Z.ai: GLM 4.5 Air
77 tok/s-2%
Context Window
256K
tokens
Larger than 58% of models
Max Output
10K
tokens
4% of context
Benchmarks
MMLU-Pro
79.3%
GPQA Diamond
72.7%
HLE
7.5%
LiveCodeBench
65.7%
SciCode
36.2%
TerminalBench Hard
17.4%
MATH-500Not evaluated
AIMENot evaluated
AIME 2025
43.3%
IFBench
41.4%
Long Context Recall
48.3%
Tau2
75.7%
Market AverageTop Score