Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works
How smart is Gemini 2.5 Pro (Jun 2025)?
Gemini 2.5 Pro (Jun 2025) is a model from Google DeepMind, released on Jun 5, 2025. Its SI-IQ is 108 (90% interval 104–112), estimated from 7 benchmark results by 3 independent evaluators. Its strongest area is Math and its weakest is Coding.
Facts
- Developer
- Google DeepMind
- Released
- Jun 5, 2025
- Access
- API access
- Benchmarks used
- 7 results from 3 evaluators
- API price
- $1.25 input / $10 output per 1M tokens (developer's price list, Oct 18, 2026)
- Context window
- 1,048,576 tokens
Ability by area (SI-IQ, 90% interval)
Every benchmark result we used
Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.
| Benchmark | Score | Evaluated by | Source |
|---|---|---|---|
| FrontierMath Tier 4 | 0.0% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| FrontierMath Tiers 1–3 | 24.6% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| OTIS Mock AIME 2024–2025 | 84.7% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| Aider Polyglot | 83.1% | Aider | Aider LLM Leaderboards · data file |
| SWE-bench Verified (Epoch AI run) | 57.6% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond | 85.3% | Epoch AI | Epoch AI Benchmarking Hub · data file |
| GPQA Diamond (OpenRouter run) | 80.6% | OpenRouter | OpenRouter benchmark datasets · data file |
| τ²-bench Airline (OpenRouter run) | 56.2% | OpenRouter | OpenRouter benchmark datasets · data file |
Human preference (Arena)
On the Arena leaderboard, where people vote between two anonymous answers, Gemini 2.5 Pro (Jun 2025) has a rating of 1,445 (1,443–1,448) from 125,366 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)
Models with similar SI-IQ
Computed with SI-IQ method version 1.0.0. Read the full methodology
Data sources: Aider polyglot leaderboard data (Apache-2.0); Arena leaderboard dataset on Hugging Face (CC BY 4.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); API list prices via LiteLLM's price table (MIT); each price links to the developer's pricing page; Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter..