JEV vs Julia: engineering task classification
We compare JEV and Julia on classifying engineering tasks in English and Russian.
Results
Results have not yet been published. No publication date has been confirmed.
| Model | Choice accuracy | Score accuracy | Noul accuracy | Overall accuracy | Latency | Cost |
|---|---|---|---|---|---|---|
| JEV | — | — | — | — | — | — |
| Julia | — | — | — | — | — | — |
Data last checked ·
How we measured
- Engineering tasks drawn from the Arcanada and Talomnia projects, in English and Russian.
- Both models receive the same requests.
- Reference answers are frozen before the models are called.
- One call per request, for each model.
- Answers are scored by deterministic rules.