HELM SeaHELM - Flores (id-en) — leaderboard
Metric: ChrF++. Source: crfm.stanford.edu. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.7 Sonnet (20250219) | 67.8 |
| 2 | GPT-4o (2024-11-20) | 67.46 |
| 3 | Llama 4 Scout Instruct | 67.19 |
| 4 | GPT-4o Mini (2024-07-18) | 66.52 |
| 5 | Qwen 2 72B Instruct | 65.8 |
| 6 | Qwen 2.5 72B Instruct | 64.08 |
| 7 | DeepSeek R1 Distill Llama 8B | 56.47 |
| 8 | Gemini 2.0 Flash | 54.27 |
| 9 | Claude 3.5 Haiku (20241022) | 49.4 |
| 10 | Llama 3.3 70B Instruct | 49.17 |
| 11 | DeepSeek V3 | 48.78 |
| 12 | Llama 4 Maverick Instruct FP8 | 48.75 |
| 13 | Llama 3.2 90B Vision Instruct | 40.04 |
| 14 | Llama 3.1 70B Instruct | 29.59 |
| 15 | Llama 3.1 405B Instruct | 11.2 |
Interactive version: theaggregate.ai/benchmark?slug=helm-seahelm-flores-id-en · How the rankings work · Data refreshed daily, snapshot 2026-07-22.