Jais 3 Model Cards - BrowseComp-Plus: leaderboard
BrowseComp-Plus deep-research browsing questions over a fixed document corpus. MBZUAI's Institute of Foundation Models reported this on the Jais 3 model cards (released October 15, 2026), comparing each Jais 3 size with size-matched reference models; this board collects the 4 cards that report it (3.7B, 7B, 32B and MoVA-36B-A4B). The cards state that all reported results use high reasoning effort. Vendor-reported. Metric: Accuracy (%). Higher is better. Source: https://huggingface.co/collections/IFM/jais-3-6abba9df4f1173ec7ce2929d
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around January 2027. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.8 27B | 53 | #104 |
| 2 | Qwen 3.5 4B | 50.6 | #473 |
| 3 | Qwen 3.5 9B | 49.5 | #377 |
| 4 | Gemma 4 31B (IT) | 42.5 | #174 |
| 5 | Muse Glimmer 30B | 41.1 | #178 |
| 6 | Gemma 4 12B (IT) | 21.9 | #371 |
| 7 | gemma-4-E4B-it | 21.2 | #566 |
| 8 | Gemma 4 E2B | 16.4 | #1076 |
| 9 | Falcon-H1-34B-Instruct | 1.8 | #666 |
| 10 | granite-4.0-h-3B-ar | 0.6 | #1133 |
| 11 | Fanar-1-9B-Instruct | 0.4 | #633 |
| 12 | Falcon-H1-7B-Instruct | 0.4 | #1043 |
| 13 | Falcon-H1-3B-Instruct | 0 | #1425 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=jais-3-model-cards-browsecomp-plus · How It Works · Data refreshed daily, snapshot 2026-10-19.