Surge AI Human Eval on Code: Results as Published

This is a blind human study of coding quality run by Mistral with Surge AI. Professional annotators rate model outputs, and the model names are hidden. Each rating is on a 1 to 5 scale. The post reports that rating as the result and does not describe the coding tasks or how the ratings are combined.

Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.