Chatbot Arena history

LMSYS Chatbot Arena Leaderboard Over Time

Explore historical LMSYS Chatbot Arena ELO rankings, compare top LLMs, replay the top 10 by date, and see which AI labs held the leaderboard. The included dataset snapshot runs through 2025-08-29.

Latest dataset leader
Gemini 2.5 pro

1,456

Aug 29, 2025

Tracked models
306

Models found in the included LMSYS dataset.

Snapshots
124

From Jan 9, 2024 to Aug 29, 2025.

Selected leader
Gemini 2.5 pro

1,456

Aug 29, 2025

LMSYS Chatbot Arena leaderboard over time
Move through historical snapshots and compare top ELO trajectories.
1,2251,3631,500Jan 9, 2024Aug 29, 2025
Gemini 2.5 pro
GPT 5 high
Claude opus 4 1 20250805 thinking 16k
o3 2025 04 16
ChatGPT 4o latest 20250326
GPT 4.5 preview 2025 02 27
GPT 5 old
Claude opus 4 1 20250805
Top 10 on Aug 29, 2025
Rank, model, company, and ELO for the selected snapshot.
RankModelCompanyELO
#1gemini-2.5-proGoogle1,456
#2gpt-5-highOpenAI1,447
#3claude-opus-4-1-20250805-thinking-16kAnthropic1,447
#4o3-2025-04-16OpenAI1,444
#5chatgpt-4o-latest-20250326OpenAI1,443
#6gpt-4.5-preview-2025-02-27OpenAI1,439
#7gpt-5-oldOpenAI1,439
#8claude-opus-4-1-20250805Anthropic1,436
#9gpt-5-chatOpenAI1,426
#10qwen-max-2025-08-15Alibaba1,425
Find a model in the dataset
Search model names included in the historical CSV.
alpaca-13b
amazon-nova-experimental-chat-05-14
amazon-nova-lite-v1.0
amazon-nova-micro-v1.0
amazon-nova-pro-v1.0
athene-70b
athene-70b-0725
athene-v2-chat
bard-jan-24-gemini-pro
c4ai-aya-expanse-32b
c4ai-aya-expanse-8b
chatglm-6b
Company leaderboard trajectory
Best model per company on each date, using the same historical snapshots.
1,0251,2631,500Jan 9, 2024Aug 29, 2025
Google
OpenAI
Anthropic
Alibaba
Longest model lead
Calendar days at rank #1 in the snapshot range.
gpt-4o-2024-05-1378d
o1-preview58d
gemini-2.5-pro50d
o1-2024-12-1742d
gemini-2.5-pro-preview-05-0629d
gpt-4-1106-preview28d
gpt-4-turbo-2024-04-0928d
gemini-2.5-pro-exp-03-2524d
Longest company lead
Which AI labs held the top spot the longest.
OpenAI324d
Google141d
Anthropic16d
Biggest leader jumps
Largest ELO gains by the #1 model between snapshots.
gemini-2.5-pro-preview-05-06+62.3

May 19, 2025 from gemini-2.5-pro-preview-05-06

gpt-4o-2024-05-13+31.6

May 15, 2024 from gpt-4-turbo-2024-04-09

gpt-4.5-preview-2025-02-27+28.6

Mar 3, 2025 from chatgpt-4o-latest-20250129

gemini-2.5-pro-exp-03-25+27.0

Mar 25, 2025 from gpt-4.5-preview-2025-02-27

o1-preview+25.7

Sep 17, 2024 from chatgpt-4o-latest-20240903

gemini-2.5-pro-preview-06-05+24.4

Jun 5, 2025 from gemini-2.5-pro-preview-05-06

gpt-5-high+22.8

Aug 8, 2025 from gemini-2.5-pro

chatgpt-4o-latest-20250129+17.6

Feb 11, 2025 from o1-2024-12-17

This page uses the included historical LMSYS Chatbot Arena demo dataset, not a live API. For the official live ranking, visit LM Arena leaderboard .

Is this LMSYS Chatbot Arena leaderboard live?

No. This tool uses the included historical demo dataset and labels the latest available snapshot date. Use the official LM Arena leaderboard for live rankings.

What is ELO in Chatbot Arena?

ELO is a rating system used to estimate relative model strength from pairwise human preference battles. Higher ELO means stronger performance in the arena dataset.

Why do some model lines appear only part of the time?

The chart focuses on models that were in the daily top 10. A model appears only on dates where it was present in that top-10 slice.

Related Calculators

P-Value Calculator

Calculate statistical significance for various tests.

Standard Deviation Calculator

Calculate population and sample standard deviation, variance, and mean.

InvNorm Calculator

Find z-scores, critical values, and x-values from normal probabilities.

Percentage Calculator

Calculate percentage of a number, percentage change, and more.

Share This Calculator

Found this tool helpful?

Help others discover it with one click.

Copy Link

Share on Social Media

XFacebookLinkedInWhatsApp
Share via Email

Suggested hashtags: #lmsys #chatbot #Calculator #FreeTools #AICalculator