gilinachum/llm.hebrew.classification.results.20230717.csv

## llm.hebrew.classification.results.20230717.csv

          
            LLM
            Accuracy
            Cost 1MT/Input
            Correct answers
            Total answers

            
              anthropic.claude-v1
              75.41%
              $11
              46
              61

            
              anthropic.claude-instant-v1
              70.49%
              $1.6
              43
              61

            
              text-davinci-003
              62.3%
              $1.5
              38
              61

            
              gpt-3.5-turbo
              50.82%
              $30
              31
              61

            
              GPT-4
              not yet tested
              $30
              -
              61

            
              anthropic.claude-v2
              not yet tested
              $11.02
              -
              61
LLM	Accuracy	Cost 1MT/Input	Correct answers	Total answers
anthropic.claude-v1	75.41%	$11	46	61
anthropic.claude-instant-v1	70.49%	$1.6	43	61
text-davinci-003	62.3%	$1.5	38	61
gpt-3.5-turbo	50.82%	$30	31	61
GPT-4	not yet tested	$30	-	61
anthropic.claude-v2	not yet tested	$11.02	-	61