Grok 3 Beta
xAILLMs
88.2
Performance
★ 4.5
Rating
145
Reviews
Text GenerationSmallMultimodalReasoningProprietaryLarge
About
xAI's most powerful model trained with 10x the compute of Grok 2, featuring a 1M token context and advanced reasoning modes.
Strengths
Trained with 10x compute of Grok 2 on the Colossus supercluster. 1M token context. Strong on MMLU-Pro, GPQA Diamond, and AIME 2025. Features Think and Big Brain reasoning modes for complex problems.
Specifications
- Context window
- 1,000,000
- Parameters
- —
Pricing
- Input cost
- $3.00 / 1M
- Output cost
- $15.00 / 1M
$3.00 input / $15.00 output per 1M tokens
Speed & Latency
- 60
- tokens/sec
- 400ms
- time to first token
- 60 RPM (free) / 600 RPM (paid)
- rate limits
Available On
xAI APIX (Twitter)Grok Web
Features
function callingstreamingvisionsystem messagesreasoning modesweb search
Performance Trend
Benchmark score trends over time for the top 5 benchmarks.
Loading history...
Benchmarks
Scores from various benchmark tests; higher is better.
| Test | Score | Percentile | Source |
|---|---|---|---|
| Chatbot Arena ELO | 1168.0 | — | chatbot-arena |
| MMLU | 79.9 | p90 | seed |