Atlas

Benchmarks

← All benchmarks

Search Arena

Search · 2025-03-18

Arena-style leaderboard for LLMs with integrated web search, reported as arena/Elo-style model scores.

Top models (higher is better)

ModelScore
GPT-5.6 Sol1261
Claude Fable 5.11246
Claude Opus 51242
GPT-6 Astra1234
GPT-6.1 Sol1233
Opus 4.71230
Claude Fable 51229
ERNIE 5.11227
Grok 4.51216
Claude Opus 5.51213
Opus 4.81204
Gemini 3 Flash Preview1196
GPT-6 Sol1175
Grok 4 Fast1170
Grok 4.31166
GPT-6 Luna1160
GPT-51133
Loading Atlas data…