AI Model Leaderboards
The definitive ranking of AI models across LLMs, Image Generation, Video, and Audio based on industry benchmarks.
Large Language Models
SourceUpdated January 25, 2026
| Rank ↑ | Model ↕ | Provider | Score ↕ | License |
|---|---|---|---|---|
| #1 | GPT-5.2 (xhigh) | OpenAI | 51 | Proprietary |
| #2 | Claude Opus 4.5 | Anthropic | 50 | Proprietary |
| #3 | GPT-5.2 Codex (xhigh) | OpenAI | 49 | Proprietary |
| #4 | Gemini 3 Pro Preview (high) | Google | 48 | Proprietary |
| #5 | GPT-5.2 (medium) | OpenAI | 47 | Proprietary |
| #6 | Gemini 3 Flash | Google | 46 | Proprietary |
| #7 | Claude Opus 4.5 | Anthropic | 43 | Proprietary |
| #8 | Claude 4.5 Sonnet | Anthropic | 43 | Proprietary |
| #9 | GPT-5.1 Codex (high) | OpenAI | 42 | Proprietary |
| #10 | GLM-4.7 | Z AI | 42 | Proprietary |
| #11 | DeepSeek V3.2 | DeepSeek | 42 | Proprietary |
| #12 | Grok 4 | xAI | 41 | Proprietary |
| #13 | Gemini 3 Pro Preview (low) | Google | 41 | Proprietary |
| #14 | GPT-5 mini (high) | OpenAI | 41 | Proprietary |
| #15 | o3 | OpenAI | 41 | Proprietary |
| #16 | Kimi K2 Thinking | Kimi | 41 | Proprietary |
| #17 | MiniMax-M2.1 | MiniMax | 40 | Proprietary |
| #18 | MiMo-V2-Flash | Xiaomi | 39 | Proprietary |
| #19 | GPT-5.1 Codex mini (high) | OpenAI | 39 | Proprietary |
| #20 | Grok 4.1 Fast | xAI | 39 | Proprietary |
| #21 | Claude 4.5 Sonnet | Anthropic | 37 | Proprietary |
| #22 | Claude 4.5 Haiku | Anthropic | 37 | Proprietary |
| #23 | KAT-Coder-Pro V1 | KwaiKAT | 36 | Proprietary |
| #24 | Nova 2.0 Pro Preview (medium) | Amazon | 36 | Proprietary |
| #25 | Gemini 3 Flash | Google | 35 | Proprietary |
| #26 | Gemini 2.5 Pro | Google | 34 | Proprietary |
| #27 | GLM-4.7 | Z AI | 34 | Proprietary |
| #28 | DeepSeek V3.2 Speciale | DeepSeek | 34 | Proprietary |
| #29 | GPT-5.2 | OpenAI | 34 | Proprietary |
| #30 | Doubao Seed Code | ByteDance Seed | 34 | Proprietary |
| #31 | gpt-oss-120B (high) | OpenAI | 33 | Proprietary |
| #32 | Qwen3 Max Thinking | Alibaba | 32 | Proprietary |
| #33 | K-EXAONE | LG AI Research | 32 | Proprietary |
| #34 | DeepSeek V3.2 | DeepSeek | 32 | Proprietary |
| #35 | Grok 3 mini Reasoning (high) | xAI | 32 | Proprietary |
| #36 | Nova 2.0 Pro Preview (low) | Amazon | 32 | Proprietary |
| #37 | Qwen3 Max | Alibaba | 31 | Proprietary |
| #38 | Claude 4.5 Haiku | Anthropic | 31 | Proprietary |
| #39 | Kimi K2 0905 | Kimi | 31 | Proprietary |
| #40 | MiMo-V2-Flash | Xiaomi | 31 | Proprietary |
| #41 | GLM-4.7-Flash | Z AI | 30 | Proprietary |
| #42 | Nova 2.0 Lite (medium) | Amazon | 30 | Proprietary |
| #43 | Qwen3 235B A22B 2507 | Alibaba | 29 | Proprietary |
| #44 | ERNIE 5.0 Thinking Preview | Baidu | 29 | Proprietary |
| #45 | Grok Code Fast 1 | xAI | 29 | Proprietary |
| #46 | Nova 2.0 Omni (medium) | Amazon | 28 | Proprietary |
| #47 | Apriel-v1.6-15B-Thinker | ServiceNow | 28 | Proprietary |
| #48 | Qwen3 VL 235B A22B | Alibaba | 28 | Proprietary |
| #49 | Magistral Medium 1.2 | Mistral | 27 | Proprietary |
| #50 | DeepSeek R1 0528 | DeepSeek | 27 | Proprietary |
| #51 | GPT-5 nano (high) | OpenAI | 27 | Proprietary |
| #52 | Qwen3 Next 80B A3B | Alibaba | 26 | Proprietary |
| #53 | Qwen3 235B 2507 | Alibaba | 25 | Proprietary |
| #54 | Qwen3 Coder 480B | Alibaba | 25 | Proprietary |
| #55 | Qwen3 VL 32B | Alibaba | 25 | Proprietary |
| #56 | gpt-oss-20B (high) | OpenAI | 24 | Proprietary |
| #57 | NVIDIA Nemotron 3 Nano | NVIDIA | 24 | Proprietary |
| #58 | Nova 2.0 Lite (low) | Amazon | 24 | Proprietary |
| #59 | gpt-oss-120B (low) | OpenAI | 24 | Proprietary |
| #60 | HyperCLOVA X SEED Think (32B) | Naver | 24 | Proprietary |
| #61 | Grok 4.1 Fast | xAI | 24 | Proprietary |
| #62 | GLM-4.6V | Z AI | 23 | Proprietary |
| #63 | Nova 2.0 Omni (low) | Amazon | 23 | Proprietary |
| #64 | GLM-4.5-Air | Z AI | 23 | Proprietary |
| #65 | K-EXAONE | LG AI Research | 23 | Proprietary |
| #66 | Mi:dm K 2.5 Pro | Korea Telecom | 23 | Proprietary |
| #67 | Nova 2.0 Pro Preview | Amazon | 23 | Proprietary |
| #68 | Mistral Large 3 | Mistral | 23 | Open Weights |
| #69 | Magistral Small 1.2 | Mistral | 23 | Proprietary |
| #70 | Ring-1T | InclusionAI | 23 | Proprietary |
| #71 | Qwen3 30B A3B 2507 | Alibaba | 22 | Proprietary |
| #72 | INTELLECT-3 | Prime Intellect | 22 | Proprietary |
| #73 | Devstral 2 | Mistral | 22 | Proprietary |
| #74 | Hermes 4 405B | Nous Research | 22 | Proprietary |
| #75 | Gemini 2.5 Flash-Lite (Sep) | Google | 22 | Proprietary |
| #76 | Solar Open 100B | Upstage | 22 | Proprietary |
| #77 | Mistral Medium 3.1 | Mistral | 21 | Open Weights |
| #78 | gpt-oss-20B (low) | OpenAI | 21 | Proprietary |
| #79 | K2-V2 (high) | MBZUAI Institute of Foundation Models | 21 | Proprietary |
| #80 | Ring-flash-2.0 | InclusionAI | 21 | Proprietary |
| #81 | Qwen3 VL 235B A22B | Alibaba | 21 | Proprietary |
| #82 | Hermes 4 70B | Nous Research | 20 | Proprietary |
| #83 | Qwen3 Next 80B A3B | Alibaba | 20 | Proprietary |
| #84 | Llama Nemotron Ultra | NVIDIA | 20 | Open Weights |
| #85 | Qwen3 Coder 30B A3B | Alibaba | 20 | Proprietary |
| #86 | Qwen3 VL 30B A3B | Alibaba | 20 | Proprietary |
| #87 | Gemini 2.5 Flash-Lite (Sep) | Google | 19 | Proprietary |
| #88 | Devstral Small 2 | Mistral | 19 | Proprietary |
| #89 | Motif-2-12.7B | Motif Technologies | 19 | Proprietary |
| #90 | Ling-1T | InclusionAI | 19 | Proprietary |
| #91 | Nova Premier | Amazon | 19 | Proprietary |
| #92 | K2-V2 (medium) | MBZUAI Institute of Foundation Models | 19 | Proprietary |
| #93 | Llama Nemotron Super 49B v1.5 | NVIDIA | 19 | Open Weights |
| #94 | Qwen3 4B 2507 | Alibaba | 19 | Proprietary |
| #95 | Llama 3.3 Nemotron Super 49B | NVIDIA | 18 | Open Weights |
| #96 | Llama 4 Maverick | Meta | 18 | Open Weights |
| #97 | Nova 2.0 Lite | Amazon | 18 | Proprietary |
| #98 | Llama 3.1 405B | Meta | 17 | Open Weights |
| #99 | ERNIE 4.5 300B A47B | Baidu | 17 | Proprietary |
| #100 | Qwen3 VL 32B | Alibaba | 17 | Proprietary |
| #101 | Hermes 4 405B | Nous Research | 17 | Proprietary |
| #102 | GLM-4.6V | Z AI | 17 | Proprietary |
| #103 | Olmo 3 7B Think | Allen Institute for AI | 17 | Proprietary |
| #104 | EXAONE 4.0 32B | LG AI Research | 17 | Proprietary |
| #105 | Qwen3 VL 8B | Alibaba | 17 | Proprietary |
| #106 | Nova 2.0 Omni | Amazon | 17 | Proprietary |
| #107 | DeepSeek R1 0528 Qwen3 8B | DeepSeek | 16 | Proprietary |
| #108 | Qwen3 VL 30B A3B | Alibaba | 16 | Proprietary |
| #109 | Ministral 3 14B | Mistral | 16 | Proprietary |
| #110 | DeepSeek R1 Distill Llama 70B | DeepSeek | 16 | Open Weights |
| #111 | Falcon-H1R-7B | TII UAE | 16 | Proprietary |
| #112 | Qwen3 Omni 30B A3B | Alibaba | 16 | Proprietary |
| #113 | Ling-flash-2.0 | InclusionAI | 15 | Proprietary |
| #114 | Ling-mini-2.0 | InclusionAI | 15 | Proprietary |
| #115 | Mistral Small 3.2 | Mistral | 15 | Open Weights |
| #116 | Qwen3 30B A3B 2507 | Alibaba | 15 | Proprietary |
| #117 | Solar Pro 2 | Upstage | 15 | Proprietary |
| #118 | Qwen3 VL 4B | Alibaba | 15 | Proprietary |
| #119 | NVIDIA Nemotron Nano 12B v2 VL | NVIDIA | 15 | Proprietary |
| #120 | NVIDIA Nemotron Nano 9B V2 | NVIDIA | 15 | Proprietary |
| #121 | Ministral 3 8B | Mistral | 15 | Proprietary |
| #122 | Llama Nemotron Super 49B v1.5 | NVIDIA | 15 | Open Weights |
| #123 | K2-V2 (low) | MBZUAI Institute of Foundation Models | 14 | Proprietary |
| #124 | Llama 3.1 Nemotron Nano 4B v1.1 | NVIDIA | 14 | Open Weights |
| #125 | Kimi Linear 48B A3B Instruct | Kimi | 14 | Proprietary |
| #126 | Reka Flash 3 | Reka AI | 14 | Proprietary |
| #127 | Llama 3.3 Nemotron Super 49B | NVIDIA | 14 | Open Weights |
| #128 | Qwen3 VL 8B | Alibaba | 14 | Proprietary |
| #129 | Llama 3.3 70B | Meta | 14 | Open Weights |
| #130 | Qwen3 VL 4B | Alibaba | 14 | Proprietary |
| #131 | Hermes 4 70B | Nous Research | 14 | Proprietary |
| #132 | Solar Pro 2 | Upstage | 14 | Proprietary |
| #133 | Llama 4 Scout | Meta | 13 | Open Weights |
| #134 | Command A | Cohere | 13 | Proprietary |
| #135 | Llama 3.1 Nemotron 70B | NVIDIA | 13 | Open Weights |
| #136 | NVIDIA Nemotron 3 Nano | NVIDIA | 13 | Proprietary |
| #137 | Qwen3 4B 2507 | Alibaba | 13 | Proprietary |
| #138 | Phi-4 | Microsoft Azure | 13 | Proprietary |
| #139 | NVIDIA Nemotron Nano 9B V2 | NVIDIA | 13 | Proprietary |
| #140 | Qwen3 1.7B | Alibaba | 13 | Proprietary |
| #141 | Olmo 3.1 32B Instruct | Allen Institute for AI | 12 | Proprietary |
| #142 | R1 1776 | Perplexity | 12 | Proprietary |
| #143 | Llama 3.2 90B (Vision) | Meta | 12 | Open Weights |
| #144 | EXAONE 4.0 32B | LG AI Research | 12 | Proprietary |
| #145 | Ministral 3 3B | Mistral | 11 | Proprietary |
| #146 | Phi-4 Mini | Microsoft Azure | 11 | Proprietary |
| #147 | DeepHermes 3 - Mistral 24B | Nous Research | 11 | Open Weights |
| #148 | Llama 3.2 11B (Vision) | Meta | 11 | Open Weights |
| #149 | Granite 4.0 H Small | IBM | 11 | Proprietary |
| #150 | Qwen3 Omni 30B A3B | Alibaba | 11 | Proprietary |
| #151 | Jamba Reasoning 3B | AI21 Labs | 10 | Proprietary |
| #152 | Nova Micro | Amazon | 10 | Proprietary |
| #153 | Gemma 3 27B | Google | 10 | Open Weights |
| #154 | NVIDIA Nemotron Nano 12B v2 VL | NVIDIA | 10 | Proprietary |
| #155 | Phi-4 Multimodal | Microsoft Azure | 10 | Proprietary |
| #156 | Gemma 3n E2B | Google | 10 | Open Weights |
| #157 | Jamba 1.7 Large | AI21 Labs | 9 | Proprietary |
| #158 | Molmo 7B-D | Allen Institute for AI | 9 | Proprietary |
| #159 | Gemma 3 12B | Google | 9 | Open Weights |
| #160 | Gemma 3 1B | Google | 9 | Open Weights |
| #161 | Gemma 3 270M | Google | 8 | Open Weights |
| #162 | Exaone 4.0 1.2B | LG AI Research | 8 | Proprietary |
| #163 | Olmo 3 7B | Allen Institute for AI | 8 | Proprietary |
| #164 | LFM2.5-1.2B-Thinking | Liquid AI | 8 | Proprietary |
| #165 | Exaone 4.0 1.2B | LG AI Research | 8 | Proprietary |
| #166 | Granite 4.0 H 1B | IBM | 8 | Proprietary |
| #167 | LFM2.5-1.2B-Instruct | Liquid AI | 8 | Proprietary |
| #168 | LFM2 2.6B | Liquid AI | 8 | Proprietary |
| #169 | Granite 4.0 Micro | IBM | 8 | Proprietary |
| #170 | DeepHermes 3 - Llama-3.1 8B | Nous Research | 8 | Open Weights |
| #171 | Jamba 1.7 Mini | AI21 Labs | 7 | Proprietary |
| #172 | Granite 4.0 1B | IBM | 7 | Proprietary |
| #173 | LFM2 8B A1B | Liquid AI | 7 | Proprietary |
| #174 | Qwen3 1.7B | Alibaba | 7 | Proprietary |
| #175 | Granite 4.0 350M | IBM | 7 | Proprietary |
| #176 | Qwen3 0.6B | Alibaba | 6 | Proprietary |
| #177 | Gemma 3 4B | Google | 6 | Open Weights |
| #178 | Gemma 3n E4B | Google | 6 | Open Weights |
| #179 | LFM2.5-VL-1.6B | Liquid AI | 6 | Proprietary |
| #180 | Qwen3 0.6B | Alibaba | 6 | Proprietary |
| #181 | Granite 4.0 H 350M | IBM | 5 | Proprietary |
| #182 | DeepSeek-OCR | DeepSeek | 0 | Proprietary |
| #183 | Grok Voice Agent | xAI | 0 | Proprietary |
| #184 | Molmo2-8B | Allen Institute for AI | 0 | Proprietary |
| #185 | Olmo 3.1 32B Think | Allen Institute for AI | 0 | Proprietary |
| #186 | Cogito v2.1 | Deep Cogito | 0 | Proprietary |
| #187 | Mi:dm K 2.5 Pro Preview | Korea Telecom | 0 | Proprietary |
| #188 | Doubao-Seed-1.8 | ByteDance Seed | 0 | Proprietary |
Showing 188 of 188 models
Rankings are aggregated from Artificial Analysis and other trusted benchmarks.
Scores represent Elo ratings, Intelligence Index, or relevant quality metrics for each category.