HiEviDR-Bench (Deep Research): leaderboard
HiEviDR-Bench (UCAS, Tsinghua and others, 2026) tests multimodal deep-research reports on its Wikipedia and arXiv question sets under the iterative deep-research pipeline. A Qwen3-VL-235B-A22B judge scores each report against a hierarchical evidence graph on five 0-20 dimensions (report quality, evidence traceability, and gated citation, claim and answer scores); the score is the mean over the Wikipedia and arXiv sets of that 0 to 100 sum, higher is better.
Metric: Overall score (0-100): sum of five 0-20 dimension scores (multimodal report quality, evidence traceability, and progressively gated citation, claim and answer scores) judged by Qwen3-VL-235B-A22B-Instruct against the hierarchical evidence graph, averaged over the Wikipedia and arXiv subsets; MLLM with Deep Research: up to three rounds of question refinement with top-20 retrieval, re-ranked to 25 text chunks and 5 images; higher is better. Source: arxiv.org. Saturation forecast: Around 2033. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4.5 | 41.18 | #82 |
| 2 | GPT-5.6 Sol | 40.13 | #30 |
| 3 | Gemma 4 31B (IT) | 39.29 | #342 |
| 4 | GPT-5 | 39.09 | #167 |
| 5 | Qwen 3.5 27B | 38.54 | #359 |
| 6 | Qwen 3.5 35B A3B | 38.28 | #482 |
| 7 | GPT-5 Mini | 38.13 | #348 |
| 8 | Qwen 3.5 9B | 37.5 | #803 |
| 9 | Qwen 3.5 4B | 37.31 | #1175 |
| 10 | Qwen 3 VL 8B Instruct | 37.11 | #981 |
| 11 | gemma-4-E4B-it | 36.76 | #1469 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=hievidr-bench-deep-research · How It Works · Data refreshed daily, snapshot 2026-11-13.