xAILanguage modelSuperseded
Grok 3
Colossus-trained model that made xAI a frontier contender.
Overall Intelligence
#49
of 56
25.7 pts behind #1
0
races led
102
weeks at #1
63.0
score
- Released
- Feb 17, 2025
- 1.7 yr ago
- Context
- 131K
- 16K max output
- Price
- $3 / $15
- per 1M in/out · $6.00 blended
- Speed
- 80 tok/s
- 0.7s to first token
- Modalities
- text · image
- outputs text
- Weights
- Closed
- API and apps
The 30-second version
Colossus-trained model that made xAI a frontier contender.
Key takeaways
- 1Colossus-trained model that made xAI a frontier contender.
Where it ranks
Current snapshot · illustrative scores · click any race to watch it
Grok 3Best on each axis
Overall
Coding
52Coding+29.6 pts50Frontend & UI+162 Elo54Backend & APIs+28.1 pts52Distributed Systems+30.1 pts53Full-Stack Apps+29.4 pts53Debugging & Repair+28.6 pts52Agentic Coding+31.7 pts54Code Review & Security+27.4 pts49Data Science & SQL+24.9 pts53Mobile Apps+27.4 pts52Terminal & DevOps+34.9 pts53Competitive Programming+823
Hard Reasoning
Math
Science
Writing & Language
Multimodal
Tool Use & Agents
Overall Intelligence over time
#1 frontierGrok 3V4.1 FlashGLM-5.3
Update history
The distilled version first; open any release for the deeper breakdown.
The Grok family