xAIGrok 3Arena #105Jun 10, 2025
Grok 3 Mini
A lightweight model that thinks before responding. Fast, smart, and great for logic-based tasks that do not require deep domain knowledge. The raw thinking traces are accessible.
Context Window
131K
tokens
Max Output
—
tokens
Released
Jun 10, 2025
Arena Rank
#105
of 305 models
Output Speed
145
tokens/sec
Time to First Token
520ms
TTFT
Capabilities
👁Vision
🧠Reasoning
🔧Tool Calling
⚡Prompt Caching
🖥Computer Use
🎨Image Generation
Supported Parameters
Include Reasoning
Show reasoning tokens
Log Probs
Token probabilities
Max Tokens
Output length limit
Reasoning
Extended thinking
Response Format
JSON mode / structured output
Seed
Deterministic outputs
Stop Sequences
Custom stop tokens
structured_outputs
Temperature
Controls randomness
Tool Choice
Control tool usage
Tool Calling
Function calling support
top_logprobs
Top P
Nucleus sampling
Pricing Comparison
| Router | Input / 1M | Output / 1M | Cached Input / 1M |
|---|---|---|---|
| Requesty★ | $0.30 | $0.50 | $0.30 |
| OpenRouter | $0.30 | $0.50 | $0.07 |
| Vercel AI | $0.30 | $0.50 | — |
| Martian | $0.30 | $0.50 | $0.07 |
Benchmarks
Artificial Analysis
Intelligence IndexArtificial Analysis
64/100Coding IndexArtificial Analysis
58/100Math IndexArtificial Analysis
78/100MMLU-PROArtificial Analysis
0.788/1GPQA DiamondArtificial Analysis
0.685/1MATH-500Artificial Analysis
0.912/1AIME 2024Artificial Analysis
0.55/1Humanity's Last ExamArtificial Analysis
0.068/1LiveCodeBenchArtificial Analysis
0.578/1SciCodeArtificial Analysis
0.318/1Model IDs
Requesty
xai/grok-3-mini:lowOpenRouter
x-ai/grok-3-miniTags
reasoningtool-calling