Qwen3.8 Launch - MADQA: leaderboard
MADQA (Multimodal Agentic Document Question Answering) asks agents to search, gather and reason over visual and textual evidence in a collection of heterogeneous PDF documents to answer human-authored questions. Scores come from the multimodal table of Alibaba Qwen's Qwen3.8-Max launch post; percent, higher is better.
Metric: Score (%). Source: qwen.ai. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.8 Max | 91.8 | #34 |
| 2 | GPT-5.6 Sol | 87.8 | #22 |
| 3 | Qwen 3.7 Plus | 87.1 | #98 |
| 4 | Claude Opus 4.8 | 86.8 | #39 |
| 5 | Claude Fable 5 | 86 | #20 |
| 6 | Gemini 3.1 Pro (Preview) | 81.1 | #64 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=qwen3-8-launch-madqa · How It Works · Data refreshed daily, snapshot 2026-11-05.