Google DeepMindLanguage modelSuperseded
Gemini 2.5 Pro
The model that put Google back on top, with a then-unmatched 1M context.
Overall Intelligence
#39
of 56
18.6 pts behind #1
0
races led
441
weeks at #1
70.1
score
- Released
- Mar 25, 2025
- 1.6 yr ago
- Context
- 1M
- 65K max output
- Price
- $1.25 / $10
- per 1M in/out · $3.44 blended
- Speed
- 140 tok/s
- 10s to first token
- Modalities
- text · image · audio · video
- outputs text
- Weights
- Closed
- API and apps
The 30-second version
The 2025 Gemini that put Google firmly in the frontier race, with a 1M context and a long run atop crowd-voted leaderboards.
Key takeaways
- 1Long reign on crowd-voted leaderboards in 2025.
- 2Strong long-context and multimodal reasoning.
Where it ranks
Current snapshot · illustrative scores · click any race to watch it
2.5 ProBest on each axis
Overall
Coding
44Coding+22.8 pts40Frontend & UI+122 Elo44Backend & APIs+20.2 pts43Distributed Systems+22.8 pts47Full-Stack Apps+23.2 pts43Debugging & Repair+22.5 pts47Agentic Coding+25.5 pts44Code Review & Security+20.1 pts41Data Science & SQL+18.0 pts49Mobile Apps+22.0 pts48Terminal & DevOps+29.0 pts44Competitive Programming+628
Hard Reasoning
Math
Science
Writing & Language
Multimodal
Tool Use & Agents
Speed & Cost
Overall Intelligence over time
#1 frontier2.5 ProV4.1 FlashGLM-5.3
Update history
The distilled version first; open any release for the deeper breakdown.
The Gemini Pro family