# Question Answering on ARC Spanish (test) Source: [Question Answering on ARC Spanish (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:32:59.076Z Evaluation of model performance on the test split of the Spanish version of the Abstract Reasoning Corpus (ARC), measured by accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Question Answering | | Dataset (including any stated split) | ARC Spanish (test) | | Metric | Accuracy | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 11 of 11 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Trinity Large | 0.92 | Not recorded | 6995476c09c33900fcd6d4aa | | Qwen3 | 0.9 | Not recorded | 6995476c09c33900fcd6d4aa | | Qwen3 | 0.86 | Not recorded | 6995476c09c33900fcd6d4aa | | Datology | 0.73 | Not recorded | 6995476c09c33900fcd6d4aa | | Granite-4.0 | 0.7 | Not recorded | 6995476c09c33900fcd6d4aa | | Llama-3.1 | 0.67 | Not recorded | 6995476c09c33900fcd6d4aa | | SmolLM3 | 0.66 | Not recorded | 6995476c09c33900fcd6d4aa | | LFM2.5 | 0.66 | Not recorded | 6995476c09c33900fcd6d4aa | | Datology | 0.63 | Not recorded | 6995476c09c33900fcd6d4aa | | Llama-3.2 | 0.56 | Not recorded | 6995476c09c33900fcd6d4aa | | Llama-3.2 | 0.31 | Not recorded | 6995476c09c33900fcd6d4aa | ### Trinity Large Accuracy: 0.92. Result ID: 699ab0c55698665ec43e330e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Architecture":"MoE","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Qwen3 Accuracy: 0.9. Result ID: 699ab0c55698665ec43e330a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"8B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Qwen3 Accuracy: 0.86. Result ID: 699ab0c55698665ec43e3309. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"4B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Datology Accuracy: 0.73. Result ID: 699ab0c55698665ec43e3305. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"8B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Granite-4.0 Accuracy: 0.7. Result ID: 699ab0c55698665ec43e330c. [Reporting paper]() Recorded evaluation configuration: {"Parameters":"Micro","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Llama-3.1 Accuracy: 0.67. Result ID: 699ab0c55698665ec43e3308. [Reporting paper]() Recorded evaluation configuration: {"Parameters":"8B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### SmolLM3 Accuracy: 0.66. Result ID: 699ab0c55698665ec43e330b. [Reporting paper]() Recorded evaluation configuration: {"Parameters":"3B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### LFM2.5 Accuracy: 0.66. Result ID: 699ab0c55698665ec43e330d. [Reporting paper]() Recorded evaluation configuration: {"Parameters":"1.2B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Datology Accuracy: 0.63. Result ID: 699ab0c55698665ec43e3304. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Parameters":"3B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Llama-3.2 Accuracy: 0.56. Result ID: 699ab0c55698665ec43e3307. [Reporting paper]() Recorded evaluation configuration: {"Parameters":"3B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. ### Llama-3.2 Accuracy: 0.31. Result ID: 699ab0c55698665ec43e3306. [Reporting paper]() Recorded evaluation configuration: {"Parameters":"1B","Protocol":"Zero-shot"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.