Model rankings
115 models · 1 boards · 1 sources
Most preferred
Best open model
Best from China
Most voted
Top three by Score
gpt-4o-2024-05-13OpenAI1,262
claude-3-5-sonnet-20240620Anthropic1,259
gemini-advanced-0514Google1,251
Top three open models
llama-3.1-405b-instruct-bf16Meta · #4 overall1,251
llama-3.1-405b-instruct-fp8Meta · #5 overall1,250
mistral-large-2407Mistral · #14 overall1,229
| Rank | Change | Model | Score | 95% interval | Votes | Price $/M |
|---|---|---|---|---|---|---|
| 1 | New | llama-3.1-405b-instruct-bf16OpenMeta | 1,251 | ±7 | 5,516 | — |
| 2 | New | llama-3.1-405b-instruct-fp8OpenMeta | 1,250 | ±4 | 30,608 | — |
| 3 | ▼ 1 | mistral-large-2407OpenMistral | 1,229 | ±4 | 26,170 | — |
| 4 | ▼ 1 | athene-70b-0725OpenNexusFlow | 1,220 | ±4 | 20,300 | — |
| 5 | New | deepseek-v2.5OpenDeepSeek | 1,219 | ±8 | 4,227 | — |
| 6 | ▼ 2 | llama-3.1-70b-instructOpenMeta | 1,210 | ±5 | 25,551 | — |
| 7 | ▼ 2 | gemma-2-27b-itOpenGoogle | 1,203 | ±4 | 36,824 | — |
| 8 | ▼ 2 | jamba-1.5-largeOpenAI21 Labs | 1,201 | ±6 | 9,185 | — |
| 9 | ▼ 1 | gemma-2-9b-it-simpoOpenPrinceton | 1,197 | ±5 | 10,624 | — |
| 10 | ▼ 3 | nemotron-4-340b-instructOpenNVIDIA | 1,196 | ±4 | 20,647 | — |
| 11 | ▼ 2 | llama-3-70b-instructOpenMeta | 1,193 | ±2 | 163,861 | — |
| 12 | ▼ 2 | command-r-plus-08-2024OpenCohere | 1,192 | ±9 | 4,975 | — |
| 13 | ▼ 2 | command-r-plusOpenCohere | 1,182 | ±3 | 80,861 | — |
| 14 | ▼ 1 | qwen2-72b-instructOpenAlibaba | 1,181 | ±3 | 38,961 | — |
| 15 | ▼ 3 | deepseek-coder-v2OpenDeepSeek | 1,180 | ±4 | 15,796 | — |
| 16 | ▼ 2 | gemma-2-9b-itOpenGoogle | 1,173 | ±4 | 29,883 | — |
| 17 | ▼ 2 | command-r-08-2024OpenCohere | 1,169 | ±8 | 5,133 | — |
| 18 | ▼ 1 | jamba-1.5-miniOpenAI21 Labs | 1,156 | ±6 | 9,311 | — |
| 19 | ▼ 3 | qwen1.5-110b-chatOpenAlibaba | 1,155 | ±3 | 27,459 | — |
| 20 | ▼ 2 | mixtral-8x22b-instruct-v0.1OpenMistral | 1,149 | ±2 | 53,813 | — |
| 21 | ▼ 2 | qwen1.5-72b-chatOpenAlibaba | 1,148 | ±4 | 40,614 | — |
| 22 | ▼ 2 | command-rOpenCohere | 1,147 | ±3 | 56,361 | — |
| 23 | ▼ 2 | llama-3-8b-instructOpenMeta | 1,141 | ±2 | 111,021 | — |
| 24 | ▼ 2 | yi-1.5-34b-chatOpen01 AI | 1,132 | ±3 | 25,148 | — |
| 25 | ▼ 2 | zephyr-orpo-141b-A35b-v0.1OpenHuggingFace | 1,131 | ±8 | 4,852 | — |
| 26 | ▼ 2 | qwen1.5-32b-chatOpenAlibaba | 1,127 | ±4 | 22,728 | — |
| 27 | ▼ 2 | llama-3.1-8b-instructOpenMeta | 1,122 | ±5 | 23,349 | — |
| 28 | ▼ 2 | phi-3-medium-4k-instructOpenMicrosoft | 1,118 | ±4 | 26,149 | — |
| 29 | ▼ 2 | mixtral-8x7b-instruct-v0.1OpenMistral | 1,114 | ±0 | 76,096 | — |
| 30 | ▼ 2 | dbrx-instruct-previewOpenDatabricks | 1,113 | ±4 | 33,721 | — |
| 31 | ▼ 2 | qwen1.5-14b-chatOpenAlibaba | 1,107 | ±4 | 18,659 | — |
| 32 | ▼ 2 | wizardlm-70bOpenMicrosoft | 1,106 | ±6 | 8,387 | — |
| 33 | ▼ 1 | gemma-2-2b-itOpenGoogle | 1,104 | ±5 | 19,964 | — |
| 34 | ▼ 3 | snowflake-arctic-instructOpenSnowflake | 1,102 | ±4 | 34,173 | — |
| 35 | ▼ 2 | openchat-3.5-0106OpenOpenChat · 1 more variant | 1,101 | ±5 | 12,971 | — |
| 36 | ▼ 2 | yi-34b-chatOpen01 AI | 1,100 | ±5 | 15,926 | — |
| 37 | ▼ 1 | phi-3-small-8k-instructOpenMicrosoft | 1,098 | ±4 | 18,492 | — |
| 38 | ▼ 3 | gemma-1.1-7b-itOpenGoogle | 1,097 | ±4 | 25,060 | — |
| 39 | ▼ 2 | tulu-2-dpo-70bOpenAllenAI/UW | 1,093 | ±6 | 6,657 | — |
| 40 | ▼ 2 | vicuna-33bOpenLMSYS | 1,092 | ±4 | 22,931 | — |
| 41 | ▼ 2 | deepseek-llm-67b-chatOpenDeepSeek | 1,091 | ±6 | 4,979 | — |
| 42 | ▼ 2 | starling-lm-7b-betaOpenNexusflow | 1,091 | ±4 | 16,658 | — |
| 43 | ▼ 2 | openhermes-2.5-mistral-7bOpenNousResearch | 1,090 | ±9 | 5,091 | — |
| 44 | ▼ 2 | starling-lm-7b-alphaOpenUC Berkeley | 1,083 | ±6 | 10,412 | — |
| 45 | ▼ 2 | llama-2-70b-chatOpenMeta | 1,082 | ±3 | 39,617 | — |
| 46 | ▼ 2 | llama2-70b-steerlm-chatOpenNVIDIA | 1,072 | ±10 | 3,634 | — |
| 47 | ▼ 2 | nous-hermes-2-mixtral-8x7b-dpoOpenNousResearch | 1,071 | ±9 | 3,838 | — |
| 48 | ▼ 2 | phi-3-mini-4k-instruct-june-2024OpenMicrosoft | 1,069 | ±6 | 12,885 | — |
| 49 | ▼ 2 | mistral-7b-instruct-v0.2OpenMistral | 1,067 | ±4 | 20,048 | — |
| 50 | ▼ 2 | qwen1.5-7b-chatOpenAlibaba | 1,067 | ±7 | 4,856 | — |
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