Skip to content

Latest commit

 

History

History
59 lines (58 loc) · 8.48 KB

2025-01-10-gemini-2.0-flash-exp-instructor-results.md

File metadata and controls

59 lines (58 loc) · 8.48 KB

Benchmark Results (model=gemini-2.0-flash-exp, date=2025-01-10)

Domain Response Model Sample Response JSON
document.invoice Invoice 
❌
document.receipt Receipt 
❌
document.resume Resume 
❌
document.us-drivers-license USDriversLicense 
{
  "issuing_state": "MT",
  "license_number": "0812319684104",
  "full_name": "BRENDA LYNN",
  "first_name": "BRENDA",
  "middle_name": null,
  "last_name": "LYNN",
  "address": {
    "street": "123 MAIN STREET",
    "city": "HELENA",
    "state": "MT",
    "zip_code": "59601"
  },
  "date_of_birth": "1968-08-04",
  "gender": "F",
  "height": "5'-06\"",
  "weight": 150.0,
  "eye_color": "BRO",
  "hair_color": null,
  "issue_date": "2015-02-15",
  "expiration_date": "2023-08-04",
  "license_class": "D",
  "donor": null,
  "veteran": null
}
document.utility-bill UtilityBill 
{
  "account_number": "1234567890-1",
  "date_mailed": "2019-09-07",
  "service_for": "SPARKY JOULE",
  "service_address": {
    "street": "12345 ENERGY CT",
    "city": null,
    "state": null,
    "zip_code": null
  },
  "billing_period_start": null,
  "billing_period_end": null,
  "date_due": "2019-09-28",
  "amount_due": 88.14,
  "previous_balance": 0.0,
  "payment_received": -91.57,
  "current_charges": 88.14,
  "breakdown_of_charges": [
    {
      "description": "Current PG&E Electric Delivery Charges",
      "amount": 55.66,
      "usage": null,
      "rate": null
    },
    {
      "description": "Silicon Valley Clean Energy Electric Generation Charges",
      "amount": 32.48,
      "usage": null,
      "rate": null
    }
  ],
  "payment_options": [
    "www.pge.com/waystopay"
  ],
  "contact_information": {
    "phone": "1-800-743-5000",
    "website": "www.pge.com/MyEnergy"
  }
}
document.w2-form W2Form 
{
  "control_number": "GN1851",
  "ein": "63-0065650",
  "ssn": "554-03-0876",
  "employee_name": "Anastasia Hodges",
  "employee_address": {
    "street": "200 2nd Street NE",
    "city": "Waseca",
    "state": "MN",
    "zip_code": "56093"
  },
  "employer_name": "NORTH 312",
  "employer_address": {
    "street": "151 N Market Street",
    "city": "Wooster",
    "state": "OH",
    "zip_code": "44691"
  },
  "wages_tips_other_compensation": 23677.7,
  "federal_income_tax_withheld": 2841.32,
  "social_security_wages": 24410.0,
  "social_security_tax_withheld": 1513.42,
  "medicare_wages_and_tips": 24410.0,
  "medicare_tax_withheld": 353.95,
  "tax_year": 2020
}
aerospace.remote-sensing RemoteSensing 
{
  "description": "This satellite image shows a large port area with numerous docks, ships, and storage facilities. Adjacent to the port is a dense urban area with a grid-like street pattern. The coastline is visible with a beach area and a breakwater extending into the ocean. The image is clear with no cloud cover.",
  "objects": [
    "ships",
    "docks",
    "storage facilities",
    "urban area",
    "beach",
    "breakwater",
    "roads",
    "railway",
    "parking lots",
    "marina"
  ],
  "categories": [
    "port",
    "residential-area",
    "beach",
    "commercial-area",
    "industrial-area",
    "parking-lot",
    "other"
  ],
  "is_visible": true
}
healthcare.medical-insurance-card MedicalInsuranceCard 
{
  "provider_service": null,
  "member_information": {
    "member_name": "Member Name",
    "member_id": "XY2123456789",
    "group_number": "023457"
  },
  "pharmacy_plan": {
    "rx_bin": "987654",
    "rx_pcn": null,
    "rx_grp": null,
    "pharmacy_help_desk": null
  },
  "insurance_provider": {
    "provider_name": "BlueCross BlueShield",
    "network": "PPO"
  },
  "coverage": {
    "office_visit": "$15",
    "specialist_visit": "$15",
    "urgent_care": null,
    "emergency_room": "$75",
    "inpatient_hospital": null
  }
}
retail.ecommerce-product-caption RetailEcommerceProductCaption 
{
  "description": "The product is a black Kindle Paperwhite with a 6.8\" display and adjustable warm light. It is shown being held in a person's hand with a book displayed on the screen.",
  "rating": 95,
  "name": "Kindle Paperwhite (8 GB)",
  "brand": "Amazon",
  "category": "Amazon Devices / Kindle",
  "price": "$139.99",
  "color": "Black"
}
media.tv-news TVNews 
{
  "description": "A female news anchor is reporting from a studio. The background is blurred and shows multiple screens. The chyron at the bottom of the screen displays the main headline and a secondary news item.",
  "chyron": "Biden criticises Netanyahu in an interview\n• Alan Bates: Post Office would do 'anything to hide Horizon failures'",
  "network": "BBC NEWS",
  "reporters": null
}
media.nfl-game-state NFLGameState 
{
  "description": null,
  "teams": [
    {
      "name": "GB",
      "score": 0
    },
    {
      "name": "AZ",
      "score": 7
    }
  ],
  "status": "in_progress",
  "quarter": 2,
  "clock_time": "12:12",
  "possession_team": null,
  "down": "2nd",
  "distance": 10,
  "yard_line": null,
  "network": "NBC",
  "is_shown": true
}
media.nba-game-state NBAGameState 
{
  "description": "The Miami Heat are playing against the San Antonio Spurs in Game 1 of the NBA Finals.",
  "teams": [
    {
      "name": "MIA",
      "score": 7
    },
    {
      "name": "SA",
      "score": 6
    }
  ],
  "status": "in_progress",
  "quarter": 1,
  "clock_time": "9:09",
  "shot_clock": 11,
  "network": "ESPN",
  "is_shown": true
}