Model rankings
451 models · 2 boards · 2 sources
Most preferred
Best open model
Best from China
Most voted
Top three by Score
gemini-2.5-proGoogle1,458
o3-2025-04-16OpenAI1,451
chatgpt-4o-latest-20250326OpenAI1,442
Top three open models
qwen3-235b-a22b-instruct-2507Alibaba · #5 overall1,433
kimi-k2-0711-previewMoonshot AI · #7 overall1,420
deepseek-r1-0528DeepSeek · #9 overall1,417
This week's changes
Qwen3 Coder 480B A35BIntelligence · #32New
Granite 3.3 Instruct 8BIntelligence · #130New
Gemma 3n E2B InstructIntelligence · #144New
Qwen3 235B A22B 2507 (Reasoning)Intelligence · #6Up 8
| Rank | Change | Model | Score | 95% interval | Votes | Price $/M |
|---|---|---|---|---|---|---|
| 1 | New | qwen3-235b-a22b-instruct-2507OpenAlibaba | 1,433 | ±10 | 3,386 | — |
| 2 | ▼ 1 | kimi-k2-0711-previewOpenMoonshot AI | 1,420 | ±6 | 10,934 | — |
| 3 | ▼ 1 | deepseek-r1-0528OpenDeepSeek · 1 more variant | 1,417 | ±6 | 17,934 | — |
| 4 | New | qwen3-235b-a22b-thinking-2507OpenAlibaba | 1,398 | ±11 | 2,777 | — |
| 5 | ▼ 2 | qwen3-235b-a22b-no-thinkingOpenAlibaba | 1,397 | ±5 | 24,015 | — |
| 6 | ▼ 2 | deepseek-v3-0324OpenDeepSeek · 1 more variant | 1,391 | ±5 | 26,999 | — |
| 7 | New | qwen3-coder-480b-a35b-instructOpenAlibaba | 1,380 | ±8 | 5,547 | — |
| 8 | ▼ 3 | minimax-m1OpenMiniMax | 1,374 | ±6 | 16,941 | — |
| 9 | ▼ 3 | qwen3-235b-a22bOpenAlibaba | 1,371 | ±5 | 20,118 | — |
| 10 | ▼ 3 | gemma-3-27b-itOpenGoogle | 1,363 | ±4 | 31,799 | — |
| 11 | ▼ 3 | mistral-small-2506OpenMistral | 1,349 | ±7 | 8,714 | — |
| 12 | ▼ 3 | qwen3-32bOpenAlibaba | 1,346 | ±9 | 4,074 | — |
| 13 | ▼ 3 | command-a-03-2025OpenCohere | 1,345 | ±5 | 30,558 | — |
| 14 | ▼ 3 | llama-3.1-nemotron-ultra-253b-v1OpenNVIDIA | 1,344 | ±12 | 2,656 | — |
| 15 | ▼ 3 | gemma-3-12b-itOpenGoogle | 1,340 | ±9 | 3,976 | — |
| 16 | ▼ 2 | qwq-32bOpenAlibaba · 1 more variant | 1,335 | ±5 | 21,864 | — |
| 17 | ▼ 4 | llama-3.1-405b-instruct-bf16OpenMeta | 1,334 | ±4 | 43,788 | — |
| 18 | ▼ 3 | llama-3.1-405b-instruct-fp8OpenMeta | 1,332 | ±3 | 63,038 | — |
| 19 | ▼ 3 | llama-4-maverick-17b-128e-instructOpenMeta | 1,330 | ±5 | 24,088 | — |
| 20 | ▼ 3 | qwen3-30b-a3bOpenAlibaba | 1,329 | ±5 | 19,956 | — |
| 21 | ▼ 3 | llama-3.3-nemotron-49b-super-v1OpenNVIDIA | 1,324 | ±12 | 2,371 | — |
| 22 | ▼ 3 | deepseek-v2.5-1210OpenDeepSeek · 1 more variant | 1,321 | ±8 | 7,243 | — |
| 23 | ▼ 2 | llama-3.3-70b-instructOpenMeta | 1,316 | ±4 | 51,606 | — |
| 24 | ▼ 2 | qwen-max-0919OpenAlibaba | 1,316 | ±5 | 17,432 | — |
| 25 | ▼ 5 | llama-4-scout-17b-16e-instructOpenMeta | 1,315 | ±6 | 13,727 | — |
| 26 | ▼ 3 | gemma-3n-e4b-itOpenGoogle | 1,314 | ±6 | 12,786 | — |
| 27 | ▼ 3 | athene-v2-chatOpenNexusFlow | 1,313 | ±4 | 26,074 | — |
| 28 | ▼ 3 | mistral-large-2407OpenMistral · 1 more variant | 1,313 | ±4 | 48,217 | — |
| 29 | ▼ 3 | athene-70b-0725OpenNexusFlow | 1,304 | ±5 | 20,580 | — |
| 30 | ▼ 3 | gemma-3-4b-itOpenGoogle | 1,303 | ±9 | 4,321 | — |
| 31 | ▼ 3 | qwen2.5-72b-instructOpenAlibaba | 1,301 | ±4 | 41,519 | — |
| 32 | ▼ 3 | llama-3.1-nemotron-70b-instructOpenNVIDIA | 1,297 | ±8 | 7,577 | — |
| 33 | ▼ 3 | llama-3.1-70b-instructOpenMeta | 1,294 | ±4 | 58,637 | — |
| 34 | ▼ 3 | mistral-small-3.1-24b-instruct-2503OpenMistral | 1,293 | ±6 | 13,698 | — |
| 35 | ▼ 3 | llama-3.1-tulu-3-70bOpenAi2 | 1,288 | ±10 | 3,010 | — |
| 36 | ▼ 3 | jamba-1.5-largeOpenAI21 Labs | 1,288 | ±7 | 9,125 | — |
| 37 | ▼ 3 | llama-3.1-nemotron-51b-instructOpenNVIDIA | 1,286 | ±10 | 3,889 | — |
| 38 | ▼ 3 | gemma-2-27b-itOpenGoogle | 1,285 | ±3 | 79,538 | — |
| 39 | ▼ 2 | gemma-2-9b-it-simpoOpenPrinceton | 1,279 | ±7 | 10,548 | — |
| 40 | ▼ 4 | nemotron-4-340b-instructOpenNVIDIA | 1,279 | ±5 | 20,608 | — |
| 41 | ▼ 3 | command-r-plus-08-2024OpenCohere | 1,278 | ±6 | 10,535 | — |
| 42 | ▼ 3 | llama-3-70b-instructOpenMeta | 1,275 | ±3 | 163,629 | — |
| 43 | ▼ 3 | mistral-small-24b-instruct-2501OpenMistral | 1,275 | ±6 | 15,321 | — |
| 44 | ▼ 3 | qwen2.5-coder-32b-instructOpenAlibaba | 1,271 | ±8 | 5,730 | — |
| 45 | ▼ 3 | c4ai-aya-expanse-32bOpenCohere | 1,268 | ±5 | 28,768 | — |
| 46 | ▼ 3 | command-r-plusOpenCohere | 1,264 | ±4 | 80,846 | — |
| 47 | ▼ 3 | qwen2-72b-instructOpenAlibaba | 1,264 | ±5 | 38,872 | — |
| 48 | ▼ 3 | gemma-2-9b-itOpenGoogle | 1,264 | ±4 | 57,197 | — |
| 49 | ▼ 3 | deepseek-coder-v2OpenDeepSeek | 1,263 | ±6 | 15,753 | — |
| 50 | ▼ 3 | phi-4OpenMicrosoft | 1,257 | ±4 | 25,213 | — |
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