An automatic evaluation tool for LLMs.
LMArena
community
AI & ML interests
None defined yet.
Recent Activity
View all activity
Organization Card
LMArena is an open platform for crowdsourced AI benchmarking, originally created by researchers from UC Berkeley SkyLab.
We have officially graduated from LMSYS.org!
Free chat with the best AI models at lmarena.ai, and see rankings at lmarena.ai/leaderboard.
Collections
2
spaces
8
Running
4.35k
Chatbot Arena Leaderboard
🏆
Display chatbot performance leaderboard
Running
2
Arena Hard Viewer
⚡
Browse and evaluate model judgments from benchmarks
Running
27
Llama-4-Maverick-03-26-Experimental Battles
🔥
Browse and compare model conversation outcomes
Running
Prompt Freshness
😻
Select similarity and language to filter prompts
Running
9
Category Arena Example
📚
Browse chatbot responses to compare models
Running
6
Preference Proxy Evaluations
🦀
Preference Proxy Evaluations
models
20
lmarena-ai/p2l-7b-grk-01112025
Updated
•
17
•
3
lmarena-ai/p2l-7b-grk-02222025
Updated
•
255
•
6
lmarena-ai/p2l-0.5b-bt-01132025
Updated
•
9
lmarena-ai/p2l-1.5b-bt-01132025
Updated
•
4
lmarena-ai/p2l-3b-bt-01132025
Updated
•
5
lmarena-ai/p2l-7b-bt-01132025
Updated
•
131
•
2
lmarena-ai/p2l-135m-bt-01132025
Updated
•
6
lmarena-ai/p2l-360m-bt-01132025
Updated
•
4
lmarena-ai/p2l-135m-rk-01132025
Updated
•
2
lmarena-ai/p2l-360m-rk-01132025
Updated
•
2
datasets
20
lmarena-ai/arena-hard-auto
Updated
•
245
lmarena-ai/categories-benchmark-eval
Preview
•
Updated
•
31
•
3
lmarena-ai/search-arena-v1-7k
Viewer
•
Updated
•
7k
•
1.03k
•
14
lmarena-ai/webdev-arena-preference-10k
Viewer
•
Updated
•
10.5k
•
212
•
7
lmarena-ai/repochat-arena-preference-4k
Viewer
•
Updated
•
3.84k
•
82
•
3
lmarena-ai/arena-human-preference-100k
Viewer
•
Updated
•
106k
•
520
•
39
lmarena-ai/VisionArena-Chat
Viewer
•
Updated
•
199k
•
3.04k
•
3
lmarena-ai/VisionArena-Battle
Viewer
•
Updated
•
29.8k
•
181
•
6
lmarena-ai/vision-arena-bench-v0.1
Viewer
•
Updated
•
500
•
1.2k
•
1
lmarena-ai/Llama-3-70b-battles
Viewer
•
Updated
•
1.6k
•
63
•
3