lmarena top 3
- User
lmarena top 3
- Assistant
The top 3 models on LMArena's leaderboard, based on the most recent available data from the provided search results, are:
-
DeepSeek-R1 (Elo: 1360)
A high-performance model developed by DeepSeek AI, licensed under MIT[3]. -
Gemma-3-27B-it (Elo: 1341)
A large-scale model from Gemma, though licensing details are unspecified[3]. -
DeepSeek-V3 (Elo: 1318)
Another model from DeepSeek AI, with an MMLU score of 88.5[3].
Note: The LMArena website[6] appears to be currently unavailable (503 Service Unavailable)[5][6], and the Hugging Face Space for the leaderboard also shows runtime errors[4]. The rankings above derive from OpenLM.ai's aggregation of LMArena data[3], which may not reflect the latest updates. For real-time rankings, users should monitor LMArena's official channels once the service resumes.
-
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
- gpt-4o-mini-search-previewCompare side by side →
- sonar-reasoningCompare side by side →
- sonarCompare side by side →
- gpt-4o-search-previewCompare side by side →
- sonar-reasoningCompare side by side →
- gpt-4o-search-preview-highCompare side by side →
- sonar-pro-highCompare side by side →
- gpt-4o-search-preview-highCompare side by side →
- sonarCompare side by side →
- gpt-4o-search-preview-high-locCompare side by side →
- gpt-4o-mini-search-previewCompare side by side →