ai benchmarks

Real-world AI performance metrics

Gauntlet Composite updated

Gauntlet Science updated

Gauntlet Ship updated

Gauntlet Teach updated

Gauntlet Judgment updated

Artificial Analysis data updated

Every change Artificial Analysis made to its scores and prices, newest first. Click a model to see its past scores.

2 price changes

Price changes 2

  • mediumCost per task$4.31 → $8.63+$4.31
  • Gemini 4 ArgonhighCost per task$4.97 → $9.95+$4.97

1 new release

New releases 1

  • peak 20.1one setting, $0.59 per task

1 removed release · 1 price change

Price changes 1

  • single settingCost per task$2.21 → $2.19−$0.021

Removed releases 1

  • was 4.8 peakhad one setting

3 price changes

Price changes 3

  • lowCost per task$4.68 → $7.19+$2.50
  • Gemini 4 CarbonmediumCost per task$4.16 → $6.59+$2.43
  • Gemini 4 CarbonhighCost per task$4.90 → $7.76+$2.86

1 new release · 3 price changes

New releases 1

  • peak 61.44 effort levels, $0.75–$6.44 per task

Price changes 3

  • lowCost per task$7.19 → $4.68−$2.50
  • Gemini 4 CarbonmediumCost per task$6.59 → $4.16−$2.43
  • Gemini 4 CarbonhighCost per task$7.76 → $4.90−$2.86

3 price changes

Price changes 3

  • lowCost per task$5.22 → $7.19+$1.97
  • Gemini 4 CarbonmediumCost per task$4.68 → $6.59+$1.91
  • Gemini 4 CarbonhighCost per task$5.48 → $7.76+$2.27

1 new release · 2 new effort levels

New releases 1

  • peak 1.7one setting, $0.12 per task

New effort levels 2

  • highIntelligence11.0$0.62 per task
  • highIntelligence9.6$0.061 per task

1 removed release · 1 price change

Price changes 1

  • single settingCost per task$0.55 → $0.17−$0.38

Removed releases 1

  • was 1.7 peakhad one setting

2 new releases · 1 price change

New releases 2

  • peak 6.0one setting, $0.55 per task
  • peak 1.7one setting, $0.43 per task

Price changes 1

  • single settingCost per task$0.56 → $0.48−$0.076

2 new releases

New releases 2

  • peak 62.15 effort levels, $0.66–$4.00 per task
  • peak 57.35 effort levels, $0.18–$1.01 per task

1 new release

New releases 1

  • peak 4.8one setting, $0.0083 per task

1 new release

New releases 1

  • peak 42.9one setting, $3.27 per task

3 price changes

Price changes 3

  • single settingCost per task$0.52 → $0.56+$0.038
  • highCost per task$0.17 → $0.18+$0.0087
  • single settingCost per task$2.23 → $2.21−$0.015

2 new effort levels

New effort levels 2

  • mediumIntelligence47.6$4.31 per task
  • highIntelligence20.8$0.85 per task

1 price change

Price changes 1

  • mediumCost per task$6.59 → $4.68−$1.91

3 price changes

Price changes 3

  • highCost per task$9.95 → $4.97−$4.97
  • lowCost per task$7.19 → $5.22−$1.97
  • Gemini 4 CarbonhighCost per task$7.76 → $5.48−$2.27

1 new release

New releases 1

  • peak 53.94 effort levels, $0.29–$3.16 per task

1 new release

New releases 1

  • peak 28.6one setting, $0.57 per task

1 new release

New releases 1

  • peak 14.2one setting, $0.96 per task

1 new release

New releases 1

  • peak 13.7one setting, $0.76 per task
Load older