AI Blending, shared and compared.
Blend multiple models, outperform single models, share your orchestra.
pip install beethovainBrowse AI Blending, shared and compared.
Blend multiple models, outperform single models, share your orchestra.
judge (ensemble)
Blending: phi4-mini[0.97] + exaone-1.2b[0.94]. 100 test cases. Duration: 570000ms
phi4-mini[0.97]exaone-1.2b[0.94]benefit_classification
@beethovain0040d
5.7ms2 models
judge (ensemble)
Blending: llama3kr-8b[0.86] + exaone-1.2b[0.94]. 100 test cases. Duration: 910000ms
llama3kr-8b[0.86]exaone-1.2b[0.94]benefit_classification
@beethovain0040d
9.1ms
judge (ensemble)
Blending: llama3kr-8b[0.86] + phi4-mini[0.97]. 100 test cases. Duration: 4660000ms
llama3kr-8b[0.86]phi4-mini[0.97]benefit_classification
@beethovain0040d
46.6ms
judge (ensemble)
Blending: exaone-7.8b[0.9] + exaone-1.2b[0.94]. 100 test cases. Duration: 1650000ms
exaone-7.8b[0.9]exaone-1.2b[0.94]benefit_classification
@beethovain0040d
16.5ms
judge (ensemble)
Blending: exaone-7.8b[0.9] + phi4-mini[0.97]. 100 test cases. Duration: 1730000ms
exaone-7.8b[0.9]phi4-mini[0.97]benefit_classification
@beethovain0040d
17.3ms
judge (ensemble)
Blending: exaone-7.8b[0.9] + llama3kr-8b[0.86]. 100 test cases. Duration: 1610000ms
exaone-7.8b[0.9]llama3kr-8b[0.86]benefit_classification
@beethovain0040d
16.1ms
judge (routing)
Blending: gemma-1b + phi4-mini. 100 test cases. Duration: 710000ms
gemma-1bphi4-minibenefit_classification
@beethovain0040d
7.1ms
judge (routing)
Blending: gemma-1b + gemma-4b. 100 test cases. Duration: 3350000ms
gemma-1bgemma-4bbenefit_classification
@beethovain0040d
33.5ms
judge (routing)
Blending: gemma-1b + llama3kr-8b. 100 test cases. Duration: 960000ms
gemma-1bllama3kr-8bbenefit_classification
@beethovain0040d
9.6ms
judge (routing)
Blending: gemma-1b + exaone-7.8b. 100 test cases. Duration: 780000ms
gemma-1bexaone-7.8bbenefit_classification
@beethovain0040d
7.8ms
judge (routing)
Blending: exaone-1.2b + phi4-mini. 100 test cases. Duration: 390000ms
exaone-1.2bphi4-minibenefit_classification
@beethovain0040d
3.9ms
judge (routing)
Blending: exaone-1.2b + gemma-4b. 100 test cases. Duration: 380000ms
exaone-1.2bgemma-4bbenefit_classification
@beethovain0040d
3.8ms
judge (routing)
Blending: exaone-1.2b + llama3kr-8b. 100 test cases. Duration: 430000ms
exaone-1.2bllama3kr-8bbenefit_classification
@beethovain0040d
4.3ms
judge (routing)
Blending: exaone-1.2b + exaone-7.8b. 100 test cases. Duration: 390000ms
exaone-1.2bexaone-7.8bbenefit_classification
@beethovain0040d
3.9ms
judge (solo)
Blending: exaone-1.2b. 100 test cases. Duration: 190000ms
exaone-1.2bbenefit_classification
@beethovain0040d
1.9ms
judge (solo)
Blending: phi4-mini. 100 test cases. Duration: 350000ms
phi4-minibenefit_classification
@beethovain0040d
3.5ms
judge (solo)
Blending: llama3kr-8b. 100 test cases. Duration: 700000ms
llama3kr-8bbenefit_classification
@beethovain0040d
7ms
judge (solo)
Blending: exaone-7.8b. 100 test cases. Duration: 800000ms
exaone-7.8bbenefit_classification
@beethovain0040d
8ms
judge (ensemble)
Blending: exaone-1.2b + gemma-1b + gemma-4b + phi4-mini. 22 test cases. Duration: 666600ms
exaone-1.2bgemma-1bgemma-4bphi4-minibenefit_classification
@beethovain0040d
judge (ensemble)
Blending: gemma-1b + gemma-4b + phi4-mini. 22 test cases. Duration: 563200ms
gemma-1bgemma-4bphi4-minibenefit_classification
@beethovain0040d
25.6ms