GoTo-AI Leaderboard - Batak Toba: leaderboard
GoTo-AI Leaderboard (GoToCompany, successor of the Sahabat-AI leaderboard), evaluating mostly open 7-72B models. Batak Toba language score from its SEA-HELM-style suite: mean of competency scores (sentiment and QA NLU, Indonesian-Batak translation both ways, instruction following), classification tasks normalized against the random baseline to 0-100; higher is better.
Metric: Score (0-100). Source: huggingface.co. Saturation forecast: Estimated already saturated. 28 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 3 27B (IT) | 52.15 | #528 |
| 2 | Llama-SEA-LION-v3-70B-IT | 41.91 | #407 |
| 3 | Llama 3.3 70B Instruct | 38.91 | #541 |
| 4 | Llama 3.1 70B Instruct | 36.42 | #567 |
| 5 | Qwen 2.5 72B Instruct | 33.65 | #469 |
| 6 | Gemma 2 9B (IT) | 32.12 | #795 |
| 7 | Gemma 2 9B | 31.08 | #924 |
| 8 | Llama 3 8B | 30.12 | #1228 |
| 9 | Llama 3.1 8B | 29.37 | #1130 |
| 10 | gemma2-9B-cpt-sahabatai-v1-instruct | 29.08 | #618 |
| 11 | DeepSeek R1 Distill Llama 70B | 28.12 | #614 |
| 12 | Llama 3 8B Instruct | 27.74 | #1116 |
| 13 | Qwen 2.5 7B | 27.07 | #895 |
| 14 | Llama 3 8B Cpt Sahabatai V1 Instruct | 26.85 | #905 |
| 15 | Qwen 2.5 7B Instruct | 26.68 | #863 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=goto-ai-leaderboard-batak-toba · How It Works · Data refreshed daily, snapshot 2026-11-05.