Technical Curator Live

Model Registry

grid_view Overview leaderboard Rankings timeline Timeline

info ELO = Battle rating from pairwise model comparisons on lmarena.ai. Higher = better. Number next to tab = Total models in the global leaderboard Arena rank = Position in the global leaderboard (not just our models)

Arena Performance · lmarena.ai 17 of 87 leaderboard models in our database Ø 1,253 Updated: Apr 1, 2026

Rank Modell Provider ELO

Ø Median 1,253

Claude Opus 4.6 Anthropic 1,476

02 #4

Claude Sonnet 4.6 Anthropic 1,389

03 #6

Kimi K2.5 Moonshot AI 1,355

04 #2

Grok 4 Fast xAI 1,338

05 #7

GLM-5 Zhipu AI 1,318

06 #3

DeepSeek-V3.2-Exp DeepSeek 1,300

07 #8

Qwen3.5-397B-A17B Alibaba / Qwen 1,295

08 #9

MiniMax M2.5 MiniMax 1,260

09 #5

GPT-5.4 mini OpenAI 1,253

Claude Opus 4.5 Anthropic 1,240

11 #6

Gemini 3.1 Pro Preview Google 1,222

12 #11

Qwen3.5-122B-A10B Alibaba / Qwen 1,220

13 #8

GPT-5.4 OpenAI 1,205

14 #12

GLM-4.6 Zhipu AI 1,195

15 #7

Claude Haiku 4.5 Anthropic 1,190

16 #13

Kimi K2-Thinking-0905 Moonshot AI 1,175

17 #14

Claude Sonnet 4.5 Anthropic 1,165

Arena Performance · lmarena.ai 8 of 75 leaderboard models in our database Ø 1,640 Updated: Apr 1, 2026

Rank Modell Provider ELO

Ø Median 1,640

Claude Opus 4.6 Anthropic 1,795

GPT-5.4 OpenAI 1,740

Gemini 3.1 Pro Preview Google 1,695

Qwen3.5-397B-A17B Alibaba / Qwen 1,660

DeepSeek-V3.2-Exp DeepSeek 1,620

Kimi K2.5 Moonshot AI 1,585

Claude Sonnet 4.6 Anthropic 1,550

GLM-5 Zhipu AI 1,510