HELM SeaHELM - Flores (vi-en) — leaderboard
Metric: ChrF++. Source: crfm.stanford.edu. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-11-20) | 61.6 |
| 2 | Claude 3.7 Sonnet (20250219) | 61.56 |
| 3 | Llama 4 Scout Instruct | 61.15 |
| 4 | Qwen 2 72B Instruct | 60.68 |
| 5 | Qwen 2.5 72B Instruct | 60.66 |
| 6 | Llama 3.3 70B Instruct | 59.61 |
| 7 | GPT-4o Mini (2024-07-18) | 59.21 |
| 8 | Llama 3.2 90B Vision Instruct | 58.75 |
| 9 | Llama 4 Maverick Instruct FP8 | 58.21 |
| 10 | Gemini 2.0 Flash | 58.04 |
| 11 | DeepSeek R1 Distill Llama 8B | 53.08 |
| 12 | Claude 3.5 Haiku (20241022) | 45.79 |
| 13 | DeepSeek V3 | 43.81 |
| 14 | Llama 3.1 70B Instruct | 28.49 |
| 15 | Llama 3.1 405B Instruct | 16.16 |
Interactive version: theaggregate.ai/benchmark?slug=helm-seahelm-flores-vi-en · How the rankings work · Data refreshed daily, snapshot 2026-07-22.