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
Google DeepMind models by SI-IQ
Google DeepMind has 34 models with independent benchmark results in our data. The best ranked is Gemini 4 Argon with an SI-IQ of 140.
Models
| 1 | Gemini 4 Argon Google DeepMind | 140137–143 | 9 |
| 2 | Gemini 3.7 Flash Google DeepMind | 131128–134 | 9 |
| 3 | Gemini 3.8 Flash Google DeepMind | 130127–133 | 9 |
| 4 | Gemini 3.1 Pro Google DeepMind | 127124–130 | 10 |
| 5 | Gemini 3.5 Flash Google DeepMind | 126123–129 | 10 |
| 6 | Gemini 3.6 Flash Google DeepMind | 125122–128 | 9 |
| 7 | Gemini 3 Flash Google DeepMind | 122119–125 | 10 |
| 8 | Gemini 3.5 Flash-Lite Google DeepMind | 109106–113 | 8 |
| 9 | Gemini 2.5 Pro (Jun 2025) Google DeepMind | 108104–112 | 8 |
| 10 | Gemma 3 27Bopen Google DeepMind | 7772–81 | 5 |
| 11 | AI Co-Mathematician Google DeepMind | 137132–142 | 1 too little data |
| 12 | Gemini 3 Pro Google DeepMind | 122115–130 | 4 too little data |
| 13 | Gemini 2.5 Pro (Mar 2025) Google DeepMind | 108102–113 | 3 too little data |
| 14 | Gemma 4 26B A4Bopen Google DeepMind | 106100–112 | 4 too little data |
| 15 | Gemini 2.5 Pro (May 2025) Google DeepMind | 10498–110 | 3 too little data |
| 16 | Gemma 4 31B ITopen Google DeepMind | 10498–110 | 5 too little data |
| 17 | Gemini 2.5 Flash (Jun 2025) Google DeepMind | 10291–113 | 3 too little data |
| 18 | Gemini 2.5 Flash (May 2025) Google DeepMind | 10294–109 | 2 too little data |
| 19 | Gemini 2.5 Flash (Apr 2025) Google DeepMind | 10194–108 | 2 too little data |
| 20 | Gemini 2.0 Pro Google DeepMind | 8983–94 | 3 too little data |
| 21 | Gemini 2.0 Flash Thinking (Jan 2025) Google DeepMind, Google | 8983–95 | 3 too little data |
| 22 | Gemini 2.0 Flash (Feb 2025) Google DeepMind, Google | 8782–91 | 3 too little data |
| 23 | Gemini 1.5 Pro (Sept 2024) Google DeepMind | 8075–85 | 3 too little data |
| 24 | Gemini 2.0 Flash (Dec 2024) Google DeepMind, Google | 8064–95 | 1 too little data |
| 25 | Gemini 1.5 Flash (Sep 2024) Google DeepMind | 7470–79 | 3 too little data |
| 26 | Gemma 3 12Bopen Google DeepMind | 7366–79 | 5 too little data |
| 27 | Gemini 1.5 Pro (May 2024) Google DeepMind | 6661–71 | 3 too little data |
| 28 | Gemma 3 4Bopen Google DeepMind | 6353–73 | 4 too little data |
| 29 | Gemini 1.5 Flash (May 2024) Google DeepMind | 5853–63 | 3 too little data |
| 30 | Gemma 2 27Bopen Google DeepMind | 5752–63 | 3 too little data |
| 31 | Gemini 1.5 Flash 8B Google DeepMind | 5747–67 | 2 too little data |
| 32 | Gemma 2 9Bopen Google DeepMind | 5347–59 | 3 too little data |
| 33 | Gemini 1.0 Pro Google DeepMind | 4640–52 | 3 too little data |
| 34 | Gemma 3 1Bopen Google DeepMind | 4025–55 | 3 too little data |
Data sources: Aider polyglot leaderboard data (Apache-2.0); Arena leaderboard dataset on Hugging Face (CC BY 4.0); Berkeley Function Calling Leaderboard, Gorilla project (Apache-2.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); LiteLLM, model_prices_and_context_window.json (MIT); Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter (licences and changes).