Model Horizon
DashboardModelsCompareBenchmarks
© 2026 Model Horizon
About|Terms
SYS.v0.1.0
Skip to content
  1. Home
  2. /Compare

Compare Models

Side-by-side comparison of AI model specs, benchmarks, and pricing.

Swap
Specifications
SpecGemini 3.1 Pro
Google
Grok 4
xAI
Tierfrontierfrontier
Release Date2026-02-192025-07-09
Context Window
1049K
256K
Max Output
66K
8K
Input Price (/1M)
$2
$3
Output Price (/1M)
$12
$15
Arena Elo
1,500
1,410
MMLU
92.6%
87.5%
GPQA
94.3%
87.5%
MATH
96.8%
91.7%
HumanEval
94.6%
90%
SWE-bench
80.6%
48%
AIME
91.2%
72%
SimpleQA
79.6%
34.2%
Capabilitiesvision, tool-use, code, reasoning, agentic, audiovision, tool-use, code, reasoning

Benchmark Comparison

Gemini 3.1 Pro
Grok 4
Arena Elo
1,500
1,410
MMLU
92.6%
87.5%
GPQA
94.3%
87.5%
MATH
96.8%
91.7%
HumanEval
94.6%
90%
SWE-bench
80.6%
48%
AIME
91.2%
72%
SimpleQA
79.6%
34.2%
Cost Estimate

Cost Calculator

Gemini 3.1 Pro
$21.00/mo$0.7000/day
Grok 4
$27.00/mo$0.9000/day
Try These Models
Try Gemini 3.1 ProTry Grok 4