Model rankings
428 models · 2 boards · 2 sources
Most preferred
Best open model
Best from China
Most voted
Top three by Score
gemini-2.5-proGoogle1,465
o3-2025-04-16OpenAI1,450
chatgpt-4o-latest-20250326OpenAI1,443
Top three open models
deepseek-r1-0528DeepSeek · #7 overall1,414
qwen3-235b-a22b-no-thinkingAlibaba · #10 overall1,399
deepseek-v3-0324DeepSeek · #13 overall1,396
| Rank | Change | Model | Score | 95% interval | Votes | Price $/M |
|---|---|---|---|---|---|---|
| 1 | — | deepseek-r1-0528OpenDeepSeek · 1 more variant | 1,414 | ±5 | 12,847 | — |
| 2 | — | qwen3-235b-a22b-no-thinkingOpenAlibaba | 1,399 | ±5 | 15,758 | — |
| 3 | — | deepseek-v3-0324OpenDeepSeek · 1 more variant | 1,396 | ±4 | 21,008 | — |
| 4 | — | minimax-m1OpenMiniMax | 1,395 | ±7 | 8,058 | — |
| 5 | — | qwen3-235b-a22bOpenAlibaba | 1,371 | ±5 | 14,597 | — |
| 6 | — | gemma-3-27b-itOpenGoogle | 1,365 | ±4 | 26,443 | — |
| 7 | New | qwen3-32bOpenAlibaba | 1,347 | ±9 | 4,074 | — |
| 8 | ▼ 1 | command-a-03-2025OpenCohere | 1,346 | ±4 | 24,524 | — |
| 9 | New | llama-3.1-nemotron-ultra-253b-v1OpenNVIDIA | 1,345 | ±11 | 2,656 | — |
| 10 | New | mistral-small-2506OpenMistral | 1,343 | ±12 | 2,061 | — |
| 11 | ▼ 3 | gemma-3-12b-itOpenGoogle | 1,342 | ±9 | 3,976 | — |
| 12 | ▼ 3 | qwq-32bOpenAlibaba · 1 more variant | 1,336 | ±4 | 18,386 | — |
| 13 | ▼ 3 | llama-3.1-405b-instruct-bf16OpenMeta | 1,334 | ±2 | 43,788 | — |
| 14 | ▼ 3 | llama-3.1-405b-instruct-fp8OpenMeta | 1,333 | ±2 | 63,038 | — |
| 15 | ▼ 3 | llama-4-maverick-17b-128e-instructOpenMeta | 1,330 | ±4 | 18,010 | — |
| 16 | New | qwen3-30b-a3bOpenAlibaba | 1,330 | ±4 | 14,229 | — |
| 17 | ▼ 4 | llama-3.3-nemotron-49b-super-v1OpenNVIDIA | 1,326 | ±11 | 2,371 | — |
| 18 | New | llama-4-scout-17b-16e-instructOpenMeta | 1,320 | ±5 | 7,451 | — |
| 19 | ▼ 5 | deepseek-v2.5-1210OpenDeepSeek · 1 more variant | 1,320 | ±6 | 7,243 | — |
| 20 | ▼ 5 | gemma-3n-e4b-itOpenGoogle | 1,315 | ±7 | 7,379 | — |
| 21 | ▼ 5 | llama-3.3-70b-instructOpenMeta | 1,315 | ±3 | 47,631 | — |
| 22 | ▼ 5 | mistral-large-2407OpenMistral · 1 more variant | 1,312 | ±2 | 48,217 | — |
| 23 | ▼ 5 | qwen-max-0919OpenAlibaba | 1,311 | ±4 | 17,432 | — |
| 24 | ▼ 5 | athene-v2-chatOpenNexusFlow | 1,311 | ±4 | 26,074 | — |
| 25 | ▼ 5 | athene-70b-0725OpenNexusFlow | 1,303 | ±3 | 20,580 | — |
| 26 | ▼ 5 | qwen2.5-72b-instructOpenAlibaba | 1,302 | ±3 | 41,519 | — |
| 27 | ▼ 5 | gemma-3-4b-itOpenGoogle | 1,301 | ±8 | 4,321 | — |
| 28 | ▼ 5 | mistral-small-3.1-24b-instruct-2503OpenMistral | 1,297 | ±8 | 7,367 | — |
| 29 | ▼ 5 | llama-3.1-nemotron-70b-instructOpenNVIDIA | 1,296 | ±7 | 7,577 | — |
| 30 | ▼ 5 | llama-3.1-70b-instructOpenMeta | 1,293 | ±2 | 58,637 | — |
| 31 | ▼ 5 | llama-3.1-tulu-3-70bOpenAi2 | 1,289 | ±9 | 3,010 | — |
| 32 | ▼ 5 | gemma-2-27b-itOpenGoogle | 1,287 | ±2 | 79,538 | — |
| 33 | ▼ 5 | jamba-1.5-largeOpenAI21 Labs | 1,285 | ±5 | 9,125 | — |
| 34 | ▼ 5 | llama-3.1-nemotron-51b-instructOpenNVIDIA | 1,284 | ±10 | 3,889 | — |
| 35 | ▼ 5 | gemma-2-9b-it-simpoOpenPrinceton | 1,280 | ±6 | 10,548 | — |
| 36 | ▼ 5 | command-r-plus-08-2024OpenCohere | 1,277 | ±5 | 10,535 | — |
| 37 | ▼ 5 | nemotron-4-340b-instructOpenNVIDIA | 1,276 | ±3 | 20,608 | — |
| 38 | ▼ 5 | llama-3-70b-instructOpenMeta | 1,275 | ±2 | 163,629 | — |
| 39 | ▼ 5 | mistral-small-24b-instruct-2501OpenMistral | 1,271 | ±5 | 15,321 | — |
| 40 | ▼ 5 | qwen2.5-coder-32b-instructOpenAlibaba | 1,270 | ±8 | 5,730 | — |
| 41 | ▼ 5 | c4ai-aya-expanse-32bOpenCohere | 1,265 | ±3 | 28,768 | — |
| 42 | ▼ 5 | command-r-plusOpenCohere | 1,263 | ±2 | 80,846 | — |
| 43 | ▼ 5 | deepseek-coder-v2OpenDeepSeek | 1,262 | ±5 | 15,753 | — |
| 44 | ▼ 5 | qwen2-72b-instructOpenAlibaba | 1,260 | ±3 | 38,872 | — |
| 45 | ▼ 5 | gemma-2-9b-itOpenGoogle | 1,260 | ±2 | 57,197 | — |
| 46 | ▼ 5 | phi-4OpenMicrosoft | 1,257 | ±4 | 25,213 | — |
| 47 | ▼ 5 | command-r-08-2024OpenCohere | 1,256 | ±5 | 10,851 | — |
| 48 | ▼ 5 | olmo-2-0325-32b-instructOpenAi2 | 1,255 | ±9 | 3,460 | — |
| 49 | ▼ 5 | jamba-1.5-miniOpenAI21 Labs | 1,240 | ±5 | 9,274 | — |
| 50 | ▼ 5 | ministral-8b-2410OpenMistral | 1,240 | ±8 | 5,111 | — |
Intelligence over time
Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi