FutureX - Level 3 (Deep Search): leaderboard

Metric: Level 3 score (latest official week, %). Source: futurex.live. 14 models tracked.

Top models

#ModelScore
1GPT-5.5 (High)67.54
2Qwen 3.7 Max57.3
3GPT-5.4 (High)44.41
4Seed 2.0 Pro44
5DeepSeek V3.2 (Thinking)35.52
6Grok 4.20 (Reasoning)35.39
7Seed 2.0 Lite33.78
8Gemini 3.1 Pro (Preview) (High)31.94
9Qwen 3.6 Max Preview27.02
10MiniMax-M2.724
11DeepSeek V4 Flash20.44
12Gemini 3.5 Flash17.33
13MiniMax-M317.31
14Grok 48.44

Interactive version: theaggregate.ai/benchmark?slug=futurex-level-3-deep-search · How It Works · Data refreshed daily, snapshot 2026-09-19.