Skip to content
IFAI

Model rankings

111 models · 1 boards · 1 sources

Artificial AnalysisIntelligence IndexArenaText arena

Text arena · Anonymous head-to-head votes · Published August 28, 2024

Most preferred
chatgpt-4o-latestOpenAI1,289Score
Best open model
Best from China
glm-4-0520Zhipu AI1,187Score
Most voted
llama-3-70b-instructMeta163,635Votes
Top three by Score
1chatgpt-4o-latestOpenAI1,289
2gpt-4o-2024-05-13OpenAI1,261
Top three open models
1llama-3.1-405b-instructMeta · #6 overall1,250
2mistral-large-2407Mistral · #14 overall1,228
3athene-70b-0725NexusFlow · #15 overall1,217
RankChangeModelScore95% intervalVotesPrice $/M
1—chatgpt-4o-latestClosedOpenAI1,289±524,023—
2—gpt-4o-2024-05-13ClosedOpenAI1,261±382,934—
3—claude-3-5-sonnet-20240620ClosedAnthropic1,258±453,352—
4—gemini-advanced-0514ClosedGoogle1,251±452,225—
5—grok-2-2024-08-13ClosedxAI1,250±610,019—
6—llama-3.1-405b-instructOpenMeta1,250±524,584—
7—gpt-4o-2024-08-06ClosedOpenAI1,245±514,886—
8—gpt-4-turbo-2024-04-09ClosedOpenAI1,240±388,005—
9—claude-3-opus-20240229ClosedAnthropic1,237±3159,592—
10—gemini-1.5-pro-api-0514ClosedGoogle1,236±374,660—
11—gpt-4-1106-previewClosedOpenAI1,234±394,283—
12—gpt-4o-mini-2024-07-18ClosedOpenAI1,231±423,147—
13—gpt-4-0125-previewClosedOpenAI1,230±387,597—
14—grok-2-mini-2024-08-13ClosedxAI1,216±610,791—
15—llama-3.1-70b-instructOpenMeta1,210±419,220—
16—gemma-2-27b-itOpenGoogle1,202±431,513—
17—gpt-4-0314ClosedOpenAI1,201±355,950—
18—gemini-1.5-flash-api-0514ClosedGoogle1,200±359,852—
19—claude-3-sonnet-20240229ClosedAnthropic1,197±2113,031—
20—nemotron-4-340b-instructOpenNVIDIA1,195±420,658—
21—llama-3-70b-instructOpenMeta1,193±2163,635—
22—gpt-4-0613ClosedOpenAI1,187±291,361—
23—claude-3-haiku-20240307ClosedAnthropic1,176±2111,796—
24—gemma-2-9b-itOpenGoogle1,173±427,150—
25—gemini-pro-dev-apiClosedGoogle1,147±518,784—
26—llama-3-8b-instructOpenMeta1,141±3109,023—
27—gpt-3.5-turbo-0125ClosedOpenAI · 1 more variant1,138±268,884—
28—gemini-proClosedGoogle1,132±86,560—
29—phi-3-medium-4k-instructOpenMicrosoft1,118±424,305—
30—llama-3.1-8b-instructOpenMeta1,117±418,335—
31—dbrx-instruct-previewOpenDatabricks1,113±433,728—
32—wizardlm-70bOpenMicrosoft1,106±78,386—
33—gemma-2-2b-itOpenGoogle1,103±614,379—
34—snowflake-arctic-instructOpenSnowflake1,102±434,184—
35—phi-3-small-8k-instructOpenMicrosoft1,097±518,493—
36—gemma-1.1-7b-itOpenGoogle1,097±425,073—
37—llama-2-70b-chatOpenMeta1,081±439,621—
38—llama2-70b-steerlm-chatOpenNVIDIA1,073±83,634—
39—phi-3-mini-4k-instruct-june-2024OpenMicrosoft1,067±712,519—
40—wizardlm-13bOpenMicrosoft1,065±67,187—
41—phi-3-mini-4k-instructOpenMicrosoft1,055±421,111—
42—llama-2-13b-chatOpenMeta1,053±519,720—
43—phi-3-mini-128k-instructOpenMicrosoft1,049±421,588—
44—codellama-34b-instructOpenMeta1,046±67,510—
45—palm-2ClosedGoogle1,046±78,744—
46—gemma-7b-itOpenGoogle1,045±69,174—
47—gemma-1.1-2b-itOpenGoogle1,036±611,356—
48—codellama-70b-instructOpenMeta1,033±191,193—
49—llama-2-7b-chatOpenMeta1,024±514,556—
50—gemma-2b-itOpenGoogle1,011±74,920—

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 “—”.