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
Models with too little data to rank
These models have fewer than 5 independent benchmark results or fewer than 2 different evaluators. Their SI-IQ is a rough estimate with a wide interval.
Provisional estimates
| 1 | GPT-5.5 Pro OpenAI | 139135–142 | 5 too little data |
| 2 | AI Co-Mathematician Google DeepMind | 137132–142 | 1 too little data |
| 3 | Muse Wonder Meta AI | 136133–139 | 4 too little data |
| 4 | GPT-5.4 Pro OpenAI | 135131–138 | 5 too little data |
| 5 | GPT-5.2 Pro OpenAI | 132128–135 | 3 too little data |
| 6 | MiMo-V2.6 Pro Xiaomi Corp | 131117–144 | 2 too little data |
| 7 | Qwen 3.8 Max Alibaba | 130127–133 | 7 too little data |
| 8 | DeepSeek-V4-Flashopen DeepSeek | 129116–143 | 2 too little data |
| 9 | Tencent Hy4 previewopen Tencent | 129115–143 | 2 too little data |
| 10 | Muse Spark 1.3 Meta AI | 128125–131 | 5 too little data |
| 11 | Qwen3.8 Max (0902) Alibaba | 127123–130 | 6 too little data |
| 12 | Muse Spark 1.1 Meta AI | 125116–133 | 2 too little data |
| 13 | Qwen3.8 Flash Alibaba | 125111–138 | 2 too little data |
| 14 | Grok 4.5 xAI | 124121–127 | 6 too little data |
| 15 | GPT-5 Pro OpenAI | 124120–128 | 3 too little data |
| 16 | Grok 4.7 xAI | 123120–127 | 7 too little data |
| 17 | Gemini 3 Pro Google DeepMind | 122115–130 | 4 too little data |
| 18 | GPT-5.3 Codex OpenAI | 122108–136 | 2 too little data |
| 19 | Mistral Large 4 Mistral AI | 122119–125 | 7 too little data |
| 20 | MiMo-V2.5-Proopen Xiaomi Corp | 121108–135 | 2 too little data |
| 21 | Muse Spark 1.2 Meta AI | 121112–129 | 2 too little data |
| 22 | MiniMax-M2.7open MiniMax | 120107–134 | 2 too little data |
| 23 | MiniMax-M2.1open MiniMax | 120106–133 | 2 too little data |
| 24 | Qwen 3.8 27Bopen Alibaba | 120106–133 | 2 too little data |
| 25 | o3-pro OpenAI | 119103–135 | 1 too little data |
| 26 | MiniMax-M2.5open MiniMax | 118105–132 | 2 too little data |
| 27 | Muse Spark Meta AI | 118110–126 | 2 too little data |
| 28 | Qwen 3.6 Max (Preview) Alibaba | 118112–124 | 5 too little data |
| 29 | MiMo-V2.5open Xiaomi Corp | 117104–131 | 2 too little data |
| 30 | Grok 4.3 Beta xAI | 117114–121 | 6 too little data |
| 31 | GLM-4.6open Z.ai (Zhipu AI), Tsinghua University | 117106–128 | 3 too little data |
| 32 | GLM-4.5-Airopen Z.ai (Zhipu AI), Tsinghua University | 11491–137 | 1 too little data |
| 33 | DeepSeek-V3.2open DeepSeek | 114101–127 | 2 too little data |
| 34 | Nemotron 3 Ultraopen Nvidia | 114108–120 | 4 too little data |
| 35 | Grok 4 xAI | 114107–120 | 4 too little data |
| 36 | Qwen 3.6 Plus Alibaba | 113109–117 | 7 too little data |
| 37 | Kimi K2 Thinkingopen Moonshot | 11299–126 | 2 too little data |
| 38 | Qwen3.5-122B-A10Bopen Alibaba | 112103–121 | 3 too little data |
| 39 | MiMo-V2.6 Flash Xiaomi Corp | 11299–125 | 2 too little data |
| 40 | DeepSeek-V3.2-Expopen DeepSeek | 112103–121 | 4 too little data |
| 41 | GLM-5open Z.ai (Zhipu AI) | 112105–118 | 4 too little data |
| 42 | Gemini 3.1 Flash-Lite Google | 109105–114 | 3 too little data |
| 43 | Qwen 3.5 Plus (hosted 397B-A17B) Alibaba | 109103–115 | 5 too little data |
| 44 | Gemini 2.5 Pro (Mar 2025) Google DeepMind | 108102–113 | 3 too little data |
| 45 | Qwen 3.6 Flash Alibaba | 107103–111 | 6 too little data |
| 46 | Qwen3.7 Flash Alibaba | 107103–112 | 4 too little data |
| 47 | GPT-5.5 Instant OpenAI | 107103–111 | 4 too little data |
| 48 | Qwen 3.5 Flash (hosted 35B-A3B) Alibaba | 107103–111 | 6 too little data |
| 49 | Qwen3-Max Alibaba | 106103–110 | 6 too little data |
| 50 | Gemma 4 26B A4Bopen Google DeepMind | 106100–112 | 3 too little data |
| 51 | Qwen3.5-35B-A3Bopen Alibaba | 106100–112 | 3 too little data |
| 52 | Claude Opus 4.1 Anthropic | 105102–109 | 6 too little data |
| 53 | Kimi K2 (Sep 2025)open Moonshot | 10592–118 | 2 too little data |
| 54 | DeepSeek-R1 (May 2025)open DeepSeek | 104100–109 | 4 too little data |
| 55 | Gemini 2.5 Pro (May 2025) Google DeepMind | 10498–110 | 3 too little data |
| 56 | Gemma 4 31B ITopen Google DeepMind | 10498–110 | 4 too little data |
| 57 | qwen3-coder-next — | 10390–116 | 2 too little data |
| 58 | o1 OpenAI | 10399–106 | 6 too little data |
| 59 | DeepSeek-V3.1-Terminusopen DeepSeek | 10289–115 | 2 too little data |
| 60 | Gemini 2.5 Flash (Jun 2025) Google DeepMind | 10291–113 | 3 too little data |
| 61 | Gemini 2.5 Flash (May 2025) Google DeepMind | 10294–109 | 2 too little data |
| 62 | Gemini 2.5 Flash (Apr 2025) Google DeepMind | 10194–108 | 2 too little data |
| 63 | Qwen3.5-9Bopen Alibaba | 10195–107 | 3 too little data |
| 64 | Grok-3 mini xAI | 10095–105 | 4 too little data |
| 65 | DeepSeek-V3.1open DeepSeek | 10087–113 | 2 too little data |
| 66 | Qwen3-30B-A3B-Thinking (Jul 2025)open Alibaba | 10093–106 | 2 too little data |
| 67 | seed-oss-36b-instruct — | 9993–106 | 2 too little data |
| 68 | DeepSeek-R1open DeepSeek | 9893–102 | 4 too little data |
| 69 | Qwen3-235B-A22B-Instruct (Jul 2025)open Alibaba | 9685–107 | 3 too little data |
| 70 | Grok 3 xAI | 9691–100 | 4 too little data |
| 71 | Qwen3.5-4Bopen Alibaba | 9688–103 | 1 too little data |
| 72 | Mistral Small 4open Mistral AI | 9683–108 | 2 too little data |
| 73 | gpt-oss-20bopen OpenAI | 9589–101 | 3 too little data |
| 74 | glm-4.7-flash — | 9488–100 | 3 too little data |
| 75 | Qwen3-30B-A3Bopen Alibaba | 9488–100 | 3 too little data |
| 76 | Cohere Command Aopen Cohere | 9375–112 | 1 too little data |
| 77 | DeepSeek-R1-Distill-Qwen-32Bopen DeepSeek | 9387–100 | 2 too little data |
| 78 | Qwen3-14Bopen Alibaba | 9387–99 | 4 too little data |
| 79 | DeepSeek-R1-Distill-Llama-70Bopen DeepSeek | 9288–97 | 3 too little data |
| 80 | QwQ-32Bopen Alibaba | 9286–98 | 3 too little data |
| 81 | o1-mini OpenAI | 9287–96 | 4 too little data |
| 82 | Qwen3-30B-A3B-Instruct (Jul 2025)open Alibaba | 9186–97 | 4 too little data |
| 83 | Qwen3-8Bopen Alibaba | 9184–97 | 3 too little data |
| 84 | deepseek-r1-0528-qwen3-8b — | 9083–98 | 2 too little data |
| 85 | QWQ-Plus Alibaba | 9079–102 | 1 too little data |
| 86 | DeepSeek-R1-Distill-Qwen-14Bopen DeepSeek | 8984–93 | 3 too little data |
| 87 | Gemini 2.0 Pro Google DeepMind | 8983–94 | 3 too little data |
| 88 | Gemini 2.0 Flash Thinking (Jan 2025) Google DeepMind, Google | 8983–95 | 3 too little data |
| 89 | GPT-4.5 OpenAI | 8783–92 | 3 too little data |
| 90 | Gemini 2.0 Flash (Feb 2025) Google DeepMind, Google | 8782–91 | 3 too little data |
| 91 | Mistral Medium 3 Mistral AI | 8681–90 | 4 too little data |
| 92 | qwen3-4b-instruct-2507 — | 8579–92 | 3 too little data |
| 93 | Qwen3-235B-A22Bopen Alibaba | 8580–91 | 3 too little data |
| 94 | DeepSeek-V3 (Mar 2025)open DeepSeek | 8581–90 | 4 too little data |
| 95 | Qwen3-Coder-480B-A35Bopen Alibaba | 8572–98 | 2 too little data |
| 96 | o1-preview OpenAI | 8480–89 | 3 too little data |
| 97 | Magistral Small 1.0open Mistral AI | 8376–89 | 2 too little data |
| 98 | Mistral Small 3.2open Mistral AI | 8074–87 | 3 too little data |
| 99 | Gemini 1.5 Pro (Sept 2024) Google DeepMind | 8075–85 | 3 too little data |
| 100 | Gemini 2.0 Flash (Dec 2024) Google DeepMind, Google | 8064–95 | 1 too little data |
| 101 | Magistral Small 1.2open Mistral AI | 7972–86 | 2 too little data |
| 102 | Qwen2.5-Max Alibaba | 7873–82 | 4 too little data |
| 103 | DeepSeek-V3open DeepSeek | 7773–82 | 4 too little data |
| 104 | glm-4.7-flash_none — | 7770–84 | 2 too little data |
| 105 | Qwen3-4Bopen Alibaba | 7765–89 | 1 too little data |
| 106 | Phi-4open Microsoft Research | 7671–81 | 4 too little data |
| 107 | Qwen Plus (Jan 2025) Alibaba | 7671–81 | 3 too little data |
| 108 | Claude 3.5 Sonnet (October 2024) Anthropic | 7671–80 | 4 too little data |
| 109 | Grok-2 (Dec 2024) xAI | 7570–80 | 3 too little data |
| 110 | Qwen2.5-Coder-32B-Instructopen Alibaba | 7559–90 | 1 too little data |
| 111 | Gemini 1.5 Flash (Sep 2024) Google DeepMind | 7470–79 | 3 too little data |
| 112 | Llama 4 Scoutopen Meta AI | 7368–78 | 4 too little data |
| 113 | Gemma 3 12Bopen Google DeepMind | 7366–79 | 4 too little data |
| 114 | DeepSeek-R1-Distill-Qwen-1.5Bopen DeepSeek | 7365–80 | 2 too little data |
| 115 | Qwen2.5-72Bopen Alibaba | 7268–77 | 4 too little data |
| 116 | Mistral Large 2 (Nov 2024)open Mistral AI | 7166–76 | 4 too little data |
| 117 | Yi-Lightning 01.AI | 7155–87 | 1 too little data |
| 118 | Claude 3.5 Sonnet Anthropic | 7166–76 | 3 too little data |
| 119 | Qwen2.5-32Bopen Alibaba | 7166–76 | 3 too little data |
| 120 | GPT-4o (May 2024) OpenAI | 7066–75 | 4 too little data |
| 121 | Llama 3.1-405Bopen Meta AI | 7065–75 | 3 too little data |
| 122 | Qwen-Turbo Alibaba | 6964–75 | 3 too little data |
| 123 | GPT-4 Turbo (Apr 2024) OpenAI | 6864–73 | 4 too little data |
| 124 | Mistral Large 2 (Jul 2024)open Mistral AI | 6863–73 | 3 too little data |
| 125 | Mistral Small 3.1open Mistral AI | 6863–73 | 3 too little data |
| 126 | openhands-lm-32b-v0.1 — | 6752–83 | 1 too little data |
| 127 | Mistral Small 3open Mistral AI | 6762–72 | 3 too little data |
| 128 | Claude 3.5 Haiku Anthropic | 6762–72 | 4 too little data |
| 129 | Gemini 1.5 Pro (May 2024) Google DeepMind | 6661–71 | 3 too little data |
| 130 | Claude 3 Opus Anthropic | 6560–70 | 5 too little data |
| 131 | Qwen3-1.7Bopen Alibaba | 6557–73 | 3 too little data |
| 132 | Qwen3-0.6Bopen Alibaba | 6445–83 | 1 too little data |
| 133 | Qwen2-72Bopen Alibaba | 6458–69 | 2 too little data |
| 134 | Llama 3.1-70Bopen Meta AI | 6358–68 | 3 too little data |
| 135 | Gemma 3 4Bopen Google DeepMind | 6353–73 | 3 too little data |
| 136 | Llama 3.2 3Bopen Meta AI | 6142–80 | 2 too little data |
| 137 | WizardLM-2 8x22Bopen Microsoft | 5953–64 | 2 too little data |
| 138 | Gemini 1.5 Flash (May 2024) Google DeepMind | 5853–63 | 3 too little data |
| 139 | Gemma 2 27Bopen Google DeepMind | 5752–63 | 3 too little data |
| 140 | GPT-4 (Jun 2023) OpenAI | 5751–63 | 4 too little data |
| 141 | Gemini 1.5 Flash 8B Google DeepMind | 5747–67 | 2 too little data |
| 142 | Llama 3-70Bopen Meta AI | 5752–62 | 3 too little data |
| 143 | Mistral Large Mistral AI | 5651–62 | 3 too little data |
| 144 | granite-4.0-1b — | 5537–74 | 2 too little data |
| 145 | Hermes 2 Theta Llama-3 70Bopen Nous Research, Arcee AI | 5550–61 | 3 too little data |
| 146 | Mixtral 8x22Bopen Mistral AI | 5549–61 | 2 too little data |
| 147 | Llama 3.1-8Bopen Meta AI | 5549–60 | 3 too little data |
| 148 | Mistral 7B v0.3open Mistral AI | 5434–75 | 2 too little data |
| 149 | granite-4.0-micro — | 5441–67 | 2 too little data |
| 150 | Claude 3 Sonnet Anthropic | 5449–59 | 3 too little data |
| 151 | Gemma 2 9Bopen Google DeepMind | 5347–59 | 3 too little data |
| 152 | granite-4.0-350m — | 5337–68 | 3 too little data |
| 153 | GPT-4 (Mar 2023) OpenAI | 5240–65 | 2 too little data |
| 154 | Claude 2.1 Anthropic | 5240–64 | 2 too little data |
| 155 | Eurus-2-7B-PRIMEopen Tsinghua University, University of Illinois Urbana-Champaign (UIUC), Shanghai AI Lab, Peking University, Shanghai Jiao Tong University, CUHK Shenzhen Research Institute | 5135–67 | 1 too little data |
| 156 | phi-3-medium 14Bopen Microsoft | 5145–57 | 2 too little data |
| 157 | Qwen2.5-7Bopen Alibaba | 5141–60 | 3 too little data |
| 158 | Claude 3 Haiku Anthropic | 5045–56 | 3 too little data |
| 159 | GPT-3.5 Turbo (Nov 2023) OpenAI | 4943–55 | 2 too little data |
| 160 | Ministral 3B Mistral AI | 4943–55 | 2 too little data |
| 161 | Claude 2 Anthropic | 4842–53 | 3 too little data |
| 162 | GPT-3.5 Turbo (Jan 2024) OpenAI | 4742–53 | 5 too little data |
| 163 | Ministral 8Bopen Mistral AI | 4741–53 | 3 too little data |
| 164 | Mistral NeMoopen Mistral AI | 4741–53 | 3 too little data |
| 165 | Gemini 1.0 Pro Google DeepMind | 4640–52 | 3 too little data |
| 166 | DBRXopen Databricks | 4640–52 | 2 too little data |
| 167 | Mixtral 8x7Bopen Mistral AI | 4337–49 | 2 too little data |
| 168 | Llama 3.2 1Bopen Meta AI | 4227–58 | 3 too little data |
| 169 | Llama 3-8Bopen Meta AI | 4034–46 | 3 too little data |
| 170 | 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); 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..