Skip to content
IFAI

Model rankings

127 models · 1 boards · 1 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published October 15, 2024

Most preferred
o1-previewOpenAI1,300Score
Best open model
Best from China
yi-lightning01 AI1,243Score
Most voted
claude-3-opus-20240229Anthropic174,238Votes
Top three by Score
1o1-previewOpenAI1,300
2gemini-1.5-pro-002Google1,271
3gpt-4o-2024-05-13OpenAI1,262
Top three open models
1llama-3.1-405b-instruct-fp8Meta · #7 overall1,252
2llama-3.1-405b-instruct-bf16Meta · #8 overall1,251
3mistral-large-2407Mistral · #19 overall1,230
RankChangeModelScore95% intervalVotesPrice $/M
1▲ 1llama-3.1-405b-instruct-fp8OpenMeta1,252±339,428—
2▼ 1llama-3.1-405b-instruct-bf16OpenMeta1,251±514,496—
3—mistral-large-2407OpenMistral1,230±435,788—
4Newqwen-max-0919OpenAlibaba1,228±512,205—
5▲ 1deepseek-v2.5OpenDeepSeek1,222±513,969—
6▼ 1qwen2.5-72b-instructOpenAlibaba1,221±513,896—
7▼ 3athene-70b-0725OpenNexusFlow1,220±420,703—
8▼ 1llama-3.1-70b-instructOpenMeta1,210±334,744—
9▼ 1gemma-2-27b-itOpenGoogle1,205±346,053—
10▼ 1jamba-1.5-largeOpenAI21 Labs1,202±69,189—
11▼ 1gemma-2-9b-it-simpoOpenPrinceton1,198±510,621—
12▼ 1nemotron-4-340b-instructOpenNVIDIA1,196±420,656—
13—command-r-plus-08-2024OpenCohere1,196±510,523—
14▼ 2llama-3-70b-instructOpenMeta1,194±2163,997—
15▼ 1command-r-plusOpenCohere1,183±280,926—
16—qwen2-72b-instructOpenAlibaba1,181±338,986—
17▼ 2deepseek-coder-v2OpenDeepSeek1,180±415,799—
18▼ 1gemma-2-9b-itOpenGoogle1,175±334,626—
19▼ 1command-r-08-2024OpenCohere1,175±610,907—
20▼ 1jamba-1.5-miniOpenAI21 Labs1,156±69,300—
21▼ 1qwen1.5-110b-chatOpenAlibaba1,155±327,476—
22▼ 1mixtral-8x22b-instruct-v0.1OpenMistral1,150±353,836—
23—qwen1.5-72b-chatOpenAlibaba1,148±340,642—
24▼ 2command-rOpenCohere1,148±356,425—
25▼ 1llama-3-8b-instructOpenMeta1,142±2109,414—
26▼ 1yi-1.5-34b-chatOpen01 AI1,133±425,147—
27▼ 1zephyr-orpo-141b-A35b-v0.1OpenHuggingFace1,131±64,862—
28▼ 1qwen1.5-32b-chatOpenAlibaba1,127±322,763—
29▼ 1llama-3.1-8b-instructOpenMeta1,125±332,937—
30▼ 1phi-3-medium-4k-instructOpenMicrosoft1,119±426,163—
31—mixtral-8x7b-instruct-v0.1OpenMistral1,114±076,163—
32—dbrx-instruct-previewOpenDatabricks1,114±333,735—
33▼ 3internlm2_5-20b-chatOpenInternLM1,112±59,176—
34▼ 1gemma-2-2b-itOpenGoogle1,110±426,098—
35▼ 1qwen1.5-14b-chatOpenAlibaba1,107±418,681—
36▼ 1wizardlm-70bOpenMicrosoft1,105±68,385—
37▼ 1snowflake-arctic-instructOpenSnowflake1,103±334,179—
38▼ 1openchat-3.5-0106OpenOpenChat · 1 more variant1,101±412,977—
39▼ 1yi-34b-chatOpen01 AI1,100±415,930—
40—phi-3-small-8k-instructOpenMicrosoft1,098±418,492—
41▼ 2gemma-1.1-7b-itOpenGoogle1,097±425,097—
42—vicuna-33bOpenLMSYS1,092±522,950—
43▲ 2starling-lm-7b-betaOpenNexusflow1,091±416,672—
44▼ 3tulu-2-dpo-70bOpenAllenAI/UW1,091±86,666—
45▲ 1llama-3.2-3b-instructOpenMeta1,090±78,620—
46▼ 3deepseek-llm-67b-chatOpenDeepSeek1,089±74,982—
47▼ 3openhermes-2.5-mistral-7bOpenNousResearch1,086±95,093—
48▼ 1starling-lm-7b-alphaOpenUC Berkeley1,083±610,417—
49▼ 1llama-2-70b-chatOpenMeta1,081±339,641—
50▼ 1llama2-70b-steerlm-chatOpenNVIDIA1,070±93,638—

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

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

Arena data comes from the leaderboard dataset it publishes on Hugging Face (CC BY 4.0). Only each board's overall leaderboard is used; scores and ranks are unchanged. dataset · CC BY 4.0

Artificial Analysis data comes from its API (artificialanalysis.ai). Scores are unchanged; ranks follow the scores. artificialanalysis.ai

Open or closed follows the license Arena gives; prices and Artificial Analysis intelligence show only where model names match, otherwise “—”.