xAI/Grok 3/Jun 10, 2025
Grok 3 Mini
A lightweight model that thinks before responding. Fast, smart, and great for logic-based tasks that don't require deep domain knowledge. Offers strong reasoning at a fraction of Grok 3's cost.
textcodereasoning
Arena ELO
1,363
Input Price
$0.3/M
Output Price
$0.5/M
Speed
183 t/s
Context
131K
Latency
720ms
Capability Assessment
No benchmark data available.
Comparative Analysis
| Metric | Grok 3 Mini | Claude Opus 4.6 | Gemini 3.1 Pro | Grok 4 |
|---|---|---|---|---|
| SWE-bench | — | 80.8% | 80.6% | 72.0% |
| AIME 2025 | 95.8% | 100.0% | 91.2% | 94.0% |
| GPQA Diamond | — | 91.3% | 94.3% | 88.0% |
| MMLU | — | 91.0% | 92.6% | — |
| Input $/M | $0.3 | $5 | $2 | $3 |
| Output $/M | $0.5 | $25 | $12 | $15 |