Jais 3 Model Cards - FLORES-200+ Translation (LLM Judge): leaderboard
FLORES-200+ translation in the cards' Arabic suite, scored by an LLM judge. 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 (7B, 32B, MoVA-36B-A4B and 375B-A23B). The cards state that all reported results use high reasoning effort. Vendor-reported. Metric: LLM-as-a-judge score. Higher is better. Source: https://huggingface.co/collections/IFM/jais-3-6abba9df4f1173ec7ce2929d
Metric: LLM-as-a-judge score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 31B (IT) | 92.8 | #174 |
| 2 | Gemma 4 12B (IT) | 89.5 | #371 |
| 3 | Qwen3.8-Flash-Next | 86.9 | #69 |
| 4 | Muse Glimmer 30B | 84.8 | #178 |
| 5 | MiniMax-M3 | 84.5 | #137 |
| 6 | gemma-4-E4B-it | 84 | #566 |
| 7 | Nemotron 3 Ultra | 83.9 | #234 |
| 8 | Qwen 3.5 397B A17B | 83.6 | #144 |
| 9 | Qwen 3.8 27B | 81.8 | #104 |
| 10 | Qwen 3.5 9B | 74.3 | #377 |
| 11 | Falcon-H1-34B-Instruct | 70.6 | #666 |
| 12 | Fanar-1-9B-Instruct | 68.4 | #633 |
| 13 | Falcon-H1-7B-Instruct | 59.8 | #1043 |
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-flores-200-plus-translation-llm-judge · How It Works · Data refreshed daily, snapshot 2026-10-19.