Skip to content
IFAI

Model rankings

121 models · 1 boards · 1 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published September 27, 2024

Most preferred
o1-previewOpenAI1,309Score
Best open model
Best from China
qwen2.5-72b-instructAlibaba1,219Score
Most voted
claude-3-opus-20240229Anthropic168,202Votes
Top three by Score
1o1-previewOpenAI1,309
2o1-miniOpenAI1,265
3gpt-4o-2024-05-13OpenAI1,262
Top three open models
1llama-3.1-405b-instruct-bf16Meta · #5 overall1,254
2llama-3.1-405b-instruct-fp8Meta · #7 overall1,251
3mistral-large-2407Mistral · #16 overall1,228
RankChangeModelScore95% intervalVotesPrice $/M
1▲ 1llama-3.1-405b-instruct-bf16OpenMeta1,254±88,787—
2▼ 1llama-3.1-405b-instruct-fp8OpenMeta1,251±433,654—
3—mistral-large-2407OpenMistral1,228±429,934—
4Newqwen2.5-72b-instructOpenAlibaba1,219±76,017—
5▼ 1athene-70b-0725OpenNexusFlow1,219±520,690—
6▼ 1deepseek-v2.5OpenDeepSeek1,218±68,362—
7▼ 1llama-3.1-70b-instructOpenMeta1,211±428,776—
8▼ 1gemma-2-27b-itOpenGoogle1,204±440,183—
9▼ 1jamba-1.5-largeOpenAI21 Labs1,202±69,178—
10▼ 1gemma-2-9b-it-simpoOpenPrinceton1,198±610,613—
11▼ 1nemotron-4-340b-instructOpenNVIDIA1,196±520,639—
12▼ 1llama-3-70b-instructOpenMeta1,194±2163,885—
13▼ 1command-r-plus-08-2024OpenCohere1,192±67,603—
14▼ 1command-r-plusOpenCohere1,183±380,872—
15▼ 1qwen2-72b-instructOpenAlibaba1,181±438,981—
16▼ 1deepseek-coder-v2OpenDeepSeek1,180±515,793—
17▼ 1gemma-2-9b-itOpenGoogle1,174±431,557—
18▼ 1command-r-08-2024OpenCohere1,170±87,838—
19—jamba-1.5-miniOpenAI21 Labs1,156±69,303—
20▼ 2qwen1.5-110b-chatOpenAlibaba1,155±427,451—
21▼ 1mixtral-8x22b-instruct-v0.1OpenMistral1,149±353,819—
22▼ 1qwen1.5-72b-chatOpenAlibaba1,148±340,607—
23▼ 1command-rOpenCohere1,148±356,353—
24▼ 1llama-3-8b-instructOpenMeta1,141±2109,319—
25▼ 1yi-1.5-34b-chatOpen01 AI1,132±425,139—
26▼ 1zephyr-orpo-141b-A35b-v0.1OpenHuggingFace1,131±74,861—
27▼ 1qwen1.5-32b-chatOpenAlibaba1,127±422,746—
28▼ 1llama-3.1-8b-instructOpenMeta1,125±426,825—
29▼ 1phi-3-medium-4k-instructOpenMicrosoft1,118±426,149—
30▼ 1mixtral-8x7b-instruct-v0.1OpenMistral1,114±076,119—
31▼ 1dbrx-instruct-previewOpenDatabricks1,113±333,701—
32—qwen1.5-14b-chatOpenAlibaba1,107±418,660—
33▲ 1gemma-2-2b-itOpenGoogle1,106±523,138—
34▼ 3wizardlm-70bOpenMicrosoft1,106±68,384—
35▼ 2snowflake-arctic-instructOpenSnowflake1,102±334,149—
36▼ 1openchat-3.5-0106OpenOpenChat · 1 more variant1,101±512,969—
37▼ 1yi-34b-chatOpen01 AI1,100±415,918—
38▼ 1phi-3-small-8k-instructOpenMicrosoft1,098±418,484—
39▼ 1gemma-1.1-7b-itOpenGoogle1,097±325,069—
40Newllama-3.2-3b-instructOpenMeta1,097±96,039—
41▼ 2tulu-2-dpo-70bOpenAllenAI/UW1,092±86,662—
42▼ 2vicuna-33bOpenLMSYS1,092±422,928—
43▼ 1starling-lm-7b-betaOpenNexusflow1,091±516,658—
44▼ 3deepseek-llm-67b-chatOpenDeepSeek1,090±84,978—
45▼ 2openhermes-2.5-mistral-7bOpenNousResearch1,089±95,090—
46Newinternlm2_5-20b-chatOpenInternLM1,086±17948—
47▼ 3starling-lm-7b-alphaOpenUC Berkeley1,083±610,414—
48▼ 3llama-2-70b-chatOpenMeta1,081±339,612—
49▼ 3llama2-70b-steerlm-chatOpenNVIDIA1,072±93,635—
50▼ 3nous-hermes-2-mixtral-8x7b-dpoOpenNousResearch1,071±103,842—

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 “—”.