A test run is a single execution of a benchmark test using a defined model configuration.
Each run represents how a particular large language model (LLM) — such as GPT-4, Claude-3, or Gemini — performed on a given task at a specific time, with specific settings.
A test run includes:
Together, test runs make it possible to compare models, providers, and configurations across benchmarks in a transparent and reproducible way.
newspaper-page data-correction 18 en
anthropic claude-sonnet-5 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<ITEM>37. <BIBL><TITLE>Eine Tossanische Bibel</TITLE></BIBL></ITEM>","number_of_corrections":1,"explanation":"The opening tag <TITLE> was malformed as \"<TITLEEine Tossanische Bibel</TITLE>\", missing the closing angle bracket \">\" after \"TITLE\". Corrected it to \"<TITLE>Eine Tossanische Bibel</TITLE>\" so the tag properly opens and closes around the text content."}newspaper-page data-correction 18 en
cohere command-r-08-2024 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{
"fixed_xml": "<?xml version=\"1.0\" encoding=\"UTF-8\"?>
<advert>
<title>The Great Adventure</title>
<author>John Smith</author>
<price>19.99</price>
<description>Embark on an epic journey with this thrilling novel. Follow the hero's quest and uncover the secrets of the ancient world.</description>
<publisher>Adventure Press</publisher>
<isbn>978-3-423-42342-3</isbn>
</advert>",
"number_of_fixes": 1,
"explanation": "The original XML was missing the root element. I added the <advert> tag to encapsulate all the content."
}newspaper-page data-correction 18 en
cohere command-r7b-12-2024 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{
"fixed_xml": "<advertisement><book title='The Great Adventure' author='John Doe' year='2023' genre='Fantasy'/></advertisement>",
"number_of_fixes": 2,
"explanation": "Added missing closing tags for the book element and ensured proper nesting. Also, added missing attributes for the book element."
}newspaper-page data-correction 18 en
cohere command-r-plus-08-2024 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<?xml version=\"1.0\" encoding=\"UTF-8\"?>
<book>
<title>The Adventures of Huckleberry Finn</title>
<author>Mark Twain</author>
<genre>Adventure</genre>
<price currency=\"USD\">12.99</price>
<description>A classic tale of a young boy's journey down the Mississippi River.</description>
</book>","number_of_corrections":1,"explanation":"Added a missing XML declaration at the beginning of the document."}newspaper-page data-correction 18 en
cohere command-a-03-2025 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<advertisement><book><title>The Great Adventure</title><author>John Doe</author><price>19.99</price><description>An exciting journey through unknown lands.</description></book><book><title>Mystery of the Lost City</title><author>Jane Smith</author><price>24.95</price><description>A thrilling mystery set in an ancient city.</description></book></advertisement>","number_of_corrections":3,"explanation":"1. Added a root element <advertisement> to encapsulate all content. 2. Added missing closing tags for <book> elements to ensure proper nesting. 3. Ensured all elements are properly closed and nested within the XML structure."}newspaper-page data-correction 18 en
genai gemini-3.1-flash-lite · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<ITEM>37. <BIBL><TITLE>Eine Tossanische Bibel</TITLE></BIBL></ITEM>","number_of_corrections":1,"explanation":"Added a missing closing bracket to the TITLE tag."}newspaper-page data-correction 18 en
x-ai grok-4.3 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<ITEM>37. <BIBL><TITLE>Eine Tossanische Bibel</TITLE></BIBL></ITEM>","number_of_corrections":1,"explanation":"Corrected malformed opening <TITLE> tag (missing '>')."}newspaper-page data-correction 18 en
openrouter meta-llama/llama-4-scout-17b-16e-instruct · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<ITEM>37. <BIBL><TITLE>Eine Tossanische Bibel</TITLE></BIBL></ITEM>","number_of_corrections":1,"explanation":"Added closing tag for TITLE element and also added a space between 'TITLE' and 'Eine' for proper XML syntax."}newspaper-page data-correction 18 en
openrouter stepfun/step-3.7-flash-20260528 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<ITEM>37. <BIBL><TITLE>Eine Tossanische Bibel</TITLE></BIBL></ITEM>","number_of_corrections":1,"explanation":"The only fault was a malformed opening <TITLE> tag missing its closing angle bracket, which created a tag name mismatch with the corresponding closing </TITLE> tag. The fix adds the missing > to correctly open the TITLE element, restoring XML well-formedness. No additional tags were added as the existing text content (the '37.' prefix) is valid as child content of the <ITEM> element."}newspaper-page data-correction 18 en
mistral mistral-large-2512 · temp 0.0 · dataclass CorrectedAdvert
One model response from this run. The per-input comparison needs JavaScript.
{"fixed_xml":"<file name=\"line_1.json\"><ITEM>37. <BIBL><TITLE>Eine Tossanische Bibel</TITLE></BIBL></ITEM></file>","number_of_corrections":1,"explanation":"Added missing closing angle bracket for the opening TITLE tag. The original XML had '<TITLEEine Tossanische Bibel</TITLE>', which was corrected to '<TITLE>Eine Tossanische Bibel</TITLE>'."}