LatamBoard
Leaderboard for model performance on Latin American and Iberian language tasks, including Spanish and Portuguese understanding, TELEIA Spanish exams, FLORES/OPUS translation, and structured or image extraction.
Metadata
Metrics
LatamBoard Category Average, Spanish Score, Spanish Spanish, Spanish Copa Es, Spanish Escola, Spanish MGSM Direct Es Spanish Bench, Spanish Openbookqa Es, Spanish Paws Es Spanish Bench, Spanish Teleia, Spanish Teleia Cervantes Ave, Spanish Teleia Pce, Spanish Teleia Siele, Spanish Wnli Es, Spanish Xnli Es Spanish Bench, Teleia Score, Teleia Teleia Cervantes Ave, Teleia Teleia Pce, Teleia Teleia Siele, Portuguese Score, Portuguese Assin2 Rte, Portuguese Bluex, Portuguese Enem Challenge, Portuguese Faquad NLI, Portuguese Oab Exams, Translation Score, Translation Flores Plus Bidirectional, Translation Flores Arb Por, Translation Flores Arb Spa, Translation Flores Cmn Por, Translation Flores Cmn Spa, Translation Flores Deu Por, Translation Flores Deu Spa, Translation Flores Eng Por, Translation Flores Eng Spa, Translation Flores Fra Por, Translation Flores Fra Spa, Translation Flores Hin Por, Translation Flores Hin Spa, Translation Flores Ita Por, Translation Flores Ita Spa, Translation Flores Por Spa, Translation Opus, Translation Opus 100 En-Es, Translation Opus 100 En-Pt, Structured Extraction Score, Structured Extraction Extraction Quality Score, Structured Extraction Composite Score, Structured Extraction Schema Validity, Structured Extraction Field F1 Partial, Structured Extraction Hallucination Rate (lower is better)
| Rank | Subject | LatamBoard Category Average | Model Match | Provenance | Sampled |
|---|---|---|---|---|---|
| 1 | surus-factura | 97.49 | — | Imported | 2026-05-06 |
| 2 | surus-extract | 91.92 | — | Imported | 2026-05-06 |
| 3 | arcee-ai/trinity-mini | 87.13 | Trinity Mini arcee-ai-trinity-mini | Imported | 2026-05-06 |
| 4 | Qwen/Qwen2.5-VL-72B-Instruct | 87.06 | Qwen2.5 VL 72B Instruct qwen-qwen2.5-vl-72b-instruct | Imported | 2026-05-06 |
| 5 | NousResearch/Hermes-4-14B | 73.72 | — | Imported | 2026-05-06 |
| 6 | Qwen/Qwen3-Next-80B-A3B-Instruct | 72.05 | Qwen3 Next 80B A3B Instruct qwen-qwen3-next-80b-a3b-instruct | Imported | 2026-05-06 |
| 7 | ibm-granite/granite-3.3-8b-instruct | 69.18 | — | Imported | 2026-05-06 |
| 8 | swiss-ai/Apertus-8B-Instruct-2509 | 68.99 | — | Imported | 2026-05-06 |
| 9 | microsoft/Phi-3-medium-4k-instruct | 68.85 | — | Imported | 2026-05-06 |
| 10 | Qwen/Qwen3-VL-8B-Instruct | 68.82 | Qwen3 VL 8B Instruct qwen-qwen3-vl-8b-instruct | Imported | 2026-05-06 |
| 11 | ibm-granite/granite-4.0-h-tiny | 68.75 | — | Imported | 2026-05-06 |
| 12 | NousResearch/Hermes-3-Llama-3.1-8B | 68.27 | — | Imported | 2026-05-06 |
| 13 | CohereLabs/aya-expanse-8b | 68.24 | — | Imported | 2026-05-06 |
| 14 | upstage/SOLAR-10.7B-Instruct-v1.0 | 68.05 | — | Imported | 2026-05-06 |
| 15 | google/gemma-3n-E4B-it | 67.31 | Gemma 3n 4B google-gemma-3n-e4b-it | Imported | 2026-05-06 |
| 16 | mistralai/Ministral-8B-Instruct-2410 | 66.21 | — | Imported | 2026-05-06 |
| 17 | mistralai/Ministral-8B-Instruct-2410 | 66.18 | — | Imported | 2026-05-06 |
| 18 | HuggingFaceH4/zephyr-7b-beta | 65.69 | — | Imported | 2026-05-06 |
| 19 | Qwen/Qwen3-4B-Instruct-2507 | 63.48 | — | Imported | 2026-05-06 |
| 20 | CohereLabs/aya-vision-8b | 63.20 | — | Imported | 2026-05-06 |
| 21 | meta-llama/Llama-3.1-8B-Instruct | 62.77 | Llama 3.1 8B Instruct meta-llama-llama-3.1-8b-instruct | Imported | 2026-05-06 |
| 22 | mann-e/Hormoz-8B | 62.77 | — | Imported | 2026-05-06 |
| 23 | Qwen/Qwen3-VL-4B-Instruct | 62.69 | — | Imported | 2026-05-06 |
| 24 | 01-ai/Yi-1.5-9B-Chat | 60.32 | — | Imported | 2026-05-06 |
| 25 | microsoft/Phi-4-multimodal-instruct | 60.28 | — | Imported | 2026-05-06 |
| 26 | google/gemma-3n-E2B-it | 59.94 | Gemma 3n 2B google-gemma-3n-e2b-it | Imported | 2026-05-06 |
| 27 | nvidia/Mistral-NeMo-Minitron-8B-Instruct | 59.88 | — | Imported | 2026-05-06 |
| 28 | nvidia/Nemotron-Mini-4B-Instruct | 57.76 | — | Imported | 2026-05-06 |
| 29 | 01-ai/Yi-1.5-6B-Chat | 56.03 | — | Imported | 2026-05-06 |
| 30 | microsoft/Phi-4-mini-instruct | 55.83 | — | Imported | 2026-05-06 |
| 31 | arcee-ai/AFM-4.5B | 55.71 | — | Imported | 2026-05-06 |
| 32 | meta-llama/Llama-3.2-3B-Instruct | 54.73 | Llama 3.2 3B Instruct meta-llama-llama-3.2-3b-instruct | Imported | 2026-05-06 |
| 33 | tencent/Hunyuan-MT-7B | 51.59 | — | Imported | 2026-05-06 |
| 34 | google/gemma-3-12b-pt | 50.37 | — | Imported | 2026-05-06 |
| 35 | deepseek-ai/DeepSeek-R1-Distill-Llama-8B | 50.13 | — | Imported | 2026-05-06 |
| 36 | deepseek-ai/DeepSeek-R1-Distill-Qwen-7B | 48.86 | — | Imported | 2026-05-06 |
| 37 | ByteDance-Seed/Seed-X-PPO-7B | 47.14 | — | Imported | 2026-05-06 |
| 38 | ByteDance-Seed/Seed-X-Instruct-7B | 46.61 | — | Imported | 2026-05-06 |
| 39 | openai/gpt-oss-20b | 38.26 | gpt-oss-20b openai-gpt-oss-20b | Imported | 2026-05-06 |
| 40 | ibm-granite/granite-vision-3.3-2b | 27.51 | — | Imported | 2026-05-06 |
| 41 | ServiceNow-AI/Apriel-1.6-15b-Thinker | 3.78 | — | Imported | 2026-05-06 |
No matching rows.