RISE Humanities Data Benchmark, 0.5.3-pre1

Search Test Runs

 

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:

  • Prompt and role definition – what the model was asked to do and from what perspective (e.g. “as a historian”).
  • Model configuration – provider, model version, temperature, and other generation parameters.
  • Results – the model’s actual response and its evaluation (scores such as F1 or accuracy).
  • Usage and cost data – token counts and calculated API costs.
  • Metadata – information like the test date, benchmark name, and person who executed it.

Together, test runs make it possible to compare models, providers, and configurations across benchmarks in a transparent and reproducible way.

Search Results
Show compact results Refine Search New Search
Download JSON Download CSV

Your search for Benchmark 'general_meeting_minutes__true' with Search Hidden 'False' returned 79 results, showing page 6 of 8.
Result 51 of 79

Test T1262 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenrouter
Modelqwen/qwen3.5-122b-a10b-20260224
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score71.40 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.71 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 28.6K IT + 29.8K OT = 58.3K TTCost: 0.007$ + 0.062$ = 0.069$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 52 of 79

Test T1265 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenrouter
Modelqwen/qwen3.5-397b-a17b-20260216
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score44.77 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.45 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 21.1K IT + 36.1K OT = 57.1K TTCost: 0.008$ + 0.084$ = 0.093$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 53 of 79

Test T1251 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenai
Modelgpt-5-nano-2025-08-07
  
Temperature1.0
DataclassMinutesPage
  
Normalized Score60.55 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.61 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 38.6K IT + 63.7K OT = 102.3K TTCost: 0.002$ + 0.025$ = 0.027$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 54 of 79

Test T1233 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Providergenai
Modelgemini-3.1-pro-preview
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score89.30 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.89 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 10.8K IT + 8.8K OT = 19.6K TTCost: 0.022$ + 0.105$ = 0.127$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 55 of 79

Test T1275 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Providerx-ai
Modelgrok-4.3
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score86.54 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.87 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 27.0K IT + 7.2K OT = 34.2K TTCost: 0.034$ + 0.018$ = 0.052$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 56 of 79

Test T1264 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenrouter
Modelqwen/qwen3.5-35b-a3b-20260224
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score79.18 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.79 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 24.7K IT + 71.5K OT = 96.2K TTCost: 0.004$ + 0.093$ = 0.097$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 57 of 79

Test T1257 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenrouter
Modelmeta-llama/llama-4-maverick-17b-128e-instruct
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score80.53 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.81 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 22.9K IT + 7.8K OT = 30.7K TTCost: 0.003$ + 0.005$ = 0.008$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 58 of 79

Test T1218 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideranthropic
Modelclaude-sonnet-5
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score67.55 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.68 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 48.9K IT + 9.0K OT = 57.9K TTCost: 0.098$ + 0.090$ = 0.188$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 59 of 79

Test T1261 at 2026-07-03

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenrouter
Modelqwen/qwen3-vl-8b-thinking
  
Temperature0.0
DataclassMinutesPage
  
Normalized Score73.74 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.74 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 28.1K IT + 94.5K OT = 122.6K TTCost: 0.003$ + 0.129$ = 0.132$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752
Result 60 of 79

Test T0707 at 2026-03-23

{'document-type': ['minutes'], 'writing': ['typed', 'handwritten'], 'century': [20], 'language': ['it', 'fr', 'de'], 'layout': ['table'], 'entry-type': ['person', 'location'], 'task': ['information-extraction']}

Configuration
Provideropenai
Modelgpt-5.3-codex
  
Temperature1.0
DataclassMinutesPage
  
Normalized Score83.63 %
Test timeunknown seconds
Prompt 

Please extract the metadata according to the given output format.
Name and Address are in the same table field. Your task is to extract them into separate fields.
Lose any dashes between name and address but preserve linebreaks.
Be aware: Addresses may contain multiple dashes; preserve them, only remove "visual splitting characters" between name and address.
Each numbered entry may span multiple lines in the name/address field.
The presented image may contain rotated pages, extract everything, treat them as a MinutesPage.
Filename: {filename}
Page: {page_number}

Results 

no valid result

Scoring
Fuzzy Score F1 micro / macro Micro precision/recall Tue/False Positives
0.84 n/a n/a n/a n/a n/a n/a n/a n/a
      Micro Precision Micro Recall Instances TP FP FN
Costs / Pricing
Pricing Date: n/a, n/a. Tokens: 21.5K IT + 5.9K OT = 27.4K TTCost: 0.038$ + 0.083$ = 0.121$
Cite: Hindermann, Marti, Alberto, et al., (2025). RISE-UNIBAS/humanities_data_benchmark, 10.5281/zenodo.16941752